VOLUME 11, ISSUE 6, JUNE 2022

Abstract: The concept of integrating information, communication, and some physical devices and being connected to a network to support the city operations and services efficiently is a part of building a smart city. In this research, IoT devices and GIS are utilized to predict the status of roads based on time to avoid traffic congestion. The GIS and IoT sensors are utilized to collect the required data to run the proposed research to control traffic lights. The info of each light is saved in a predefined list and equipped with three lights: red, orange, and green. These colours are used to indicate the status of the road in front of the drivers. The proposed method introduces a novel algorithm to avoid the traffic using GIS and IoT devices to avoid traffic congestion.

Keywords: Smart Road, Internet of Things (IoT), Traffic Congestion, Geographical Information Systems (GIS)

Abstract: Books play a very important role in every person’s life by introducing them to a world of imagination, providing knowledge of the outside world, improving their reading, writing, and speaking skills as well as boosting memory and intelligence, all of which are quite necessary for different aspects of life. There exist many potential readers, but due to the abundance of information present on the internet, many of these people find it extremely hard to search for books that they might like and might inculcate in them a habit of reading, which is always encouraged. This could result in a huge loss as a lack of readings results in poor language skills, cultural ignorance, and fear of books. Furthermore, many people ask for book recommendations from their friends, neighbors, and families who might not always suggest the right book as they do not have knowledge about the numerous books that are available. If we plan to buy any new book, we normally ask our friends, research about the book, check the book ratings on the internet, find books that have similar content, and then we make our decision. How convenient if all this process was taken care of automatically and recommend the book efficiently? A recommendation system is an answer to this question. Recommendation System (RS) is software that suggests similar items to a purchaser based on their earlier purchases or preferences. The amount of information available on the internet is quite a lot and finding relevant information can become very difficult. Recommendation systems aim to solve such kinds of problems. With the help of recommendation systems, we can find relevant information quickly and easily. Many recommendation systems are also used in commercial websites to sell their products. Consequently, the main aim of our paper is to build a book recommendation web application. The web application can be used by anyone and does not require any login making it much more accessible and easy to use. The user needs to just enter the title of the book that they have read and liked before, and based on the genre and average rating given to the book. The top ten books will be recommended to the user that is the most similar to the book that they have entered. The technologies used in this paper would be the python programming language for preprocessing the data, exploring the data, and building the machine learning model itself. Many python modules such as pandas and matplotlib and seaborn will be used for handling and visualizing the data. The algorithm that will be used to find books similar to the book entered by the user is the K-Means nearest neighbor algorithm. To provide a proper and appealing interface, this project is going to be a web application that will be developed using Flask. The initial exploration and preprocessing of the data will be done through Google Colab. The dataset used is ‘books_1.Best_Books_Ever’ from the ‘Goodreads’ dataset that contains attributes such as booke, title, author, rating, ISBN, genres, characters, awards, num ratings, ratingByStars, setting, bbeScore, bbeVotes etc.

Keywords: K-Means nearest neighbor; machine learning; Books; Receommendation;

Abstract: The importance of face recognition system in our world today is enormous. Issues that the existing face recognition systems had to solve emanated from in consistence in facial patterns which do not conform to the traditional facial patterns. Existing face recognition models which employed the use of genetic algorithm for modeling CNN have the problem of slow convergence and local minimal entrapment. Firefly algorithm (FA) has been found to produce consistent and better performance in terms of time and optimality than other algorithm. Firefly (FA) is therefore applied for modeling CNN face recognition system. Three models were used, FFA-CNN, CNN1, and CNN2. Where FFA-CNN is a CNN model designed to obtain optimized parameter using FFA as the optimizer, CNN1 and CNN2 are CNN model designed by Random model parameters. For each model a total number of 694 sample facial images which is about (70%) of total dataset were used for training and 299 sample of facial images which is about (30%) of total dataset were used for testing the trained system. 3 experiments were carried out. The result shows that FFA-CNN is 100% accurate while CNN1 and CNN2 are 30.10% and 81.61% respectively. With this excellent result, FFA-CNN model develop in this research work can be recommended for use in face recognition system. This research has contributed immensely to knowledge by developing an algorithm that improved the performance of CNN model for face recognition.

Keywords: Convolutional Neural Network, Face Recognition System, Firefly Algorithm, Optimized Parameter.

Abstract: In this research, we have developed a Smart Classrooms Management System (SCMS). The SCMS provides a comprehensive solution to classroom management through the engagement of smart devices, the Internet, and computer applications. The system consists of several components; these are classroom scheduling and access control, student attendance, and lectures management. The work has been divided into three phases; In phase one, we have developed the student attendance registration and monitoring system by using the QR Code technology. The attendance system was recorded manually. The manual process required more time, labour, and prune to human errors. The manual attendance has been replaced with automatic methods that utilizes smartphones based on QR Code technology. The work has shown promising results which made classroom management smarter and more efficient.

Keywords: Smart Campus, QR Code Technology, Internet of Things, Smart Classroom Management

Abstract: Our paper presents an IoT device which help to manage and plan the usage of water. This system can be easily installed in residential societies. Sensors placed in the tank which continuously informs the water level at the current time. This information will be updated on the cloud and using an android application, user can visualize the water level on a smartphone anywhere that is connected to Internet updated using the data collected using sensors. As we know water is so precious for human being as well as for the complete nature without which it will not be possible to survive. Even though lot many efforts have been taken by government though various schemes, it is becoming difficult day by day to save water for future and make efficient utilization of it. Hence the main focus is on water utilization in apartments and save water with proper distribution and monitoring system. The intensions of this work are water management, monitoring and system of proper distribution of water to save water and make efficient use of it, so that we can satisfy the trust of others. The system has been designed in such a way that it will monitor the available water level continuously. System has been implemented using embedded system and communication takes place through IoT.

Keywords: Water monitoring, IoT, Smartphone, Home Assisting Devices

Abstract: This project is developed by using Radio Frequency Identification (RFID) system and student card to get student attendance. Before this project, lecturers needed to use paper to get the student attendance. There were a lot of problems when using the paper as student attendance such as cheating. This project can help lecturers to reduce these problems by the design of an automatic attendance using RFID and student card. The project system runs by the process of getting the code of the student card to compare with the database in XAMPP Control Panel. Graphical User Interface (GUI) was developed using Net Beans IDE 8.1 to make the database easier to access. Firstly, lecturer needs to fill forms in an interface like lecturer name, subject and subject code. This part is important because we need the information in this part to use in the next interface. In the next interface, lecturer needs to choose port and speed to make connection with RFID reader. After the reader is ready, process to get attendant will start. Students need to swipe their card on the reader and the code from the card will use to compare with database in XAMPP Control Panel. When the code is match with database, the student information like ID number and time will show on interface and that information will trigger into a list and it will lead to the opening of the class room door. This list will use as a student attendance. In that list, all information like student name, ID number and time will be saved in the database of the server. If the code does not match with the database, it means that the student is in the wrong class or he (or she) is not yet registered for that course. When this happen, lecturer can register that student by using registering form and the information of that student will be update into database. This project will help lecturer taking the student attendance more easily and automatically. As a conclusion, RFID technology can be used in student attendance application.

Keywords: RFID MODULE, LCD , MICROCONTROLLER, COMPUTER SYSTEM

Abstract: Analysis of public information from social media could yield interesting results and insights into the globe of public opinions about almost any product, service or personality. Social network data is one amongst the foremost effective and accurate indicators of public sentiment. The explosion of Web 2.0 has led to increased activity in Podcasting, Blogging, Tagging, Contributing to RSS, Social Bookmarking, and Social Networking. As a result there has been an eruption of interest in people to mine these vast resources of knowledge for opinions. Sentiment Analysis or Opinion Mining is that the computational treatment of opinions, sentiments and subjectivity of text. during this paper we are going to be discussing a strategy which allows utilization and interpretation of twitter data to see public opinions. Developing a program for sentiment analysis is an approach to be accustomed computationally measure customers' perceptions. This paper reports on the look of a sentiment analysis, extracting and training an unlimited amount of datasets. Results classify customers' perspective via datasets into positive and negative, which is represented during a chart, bar diagram, scatter plot using php, css and html pages.

Keywords: data processing, linguistic communication processing, Naïve Bayes.

Abstract: The data available online, helps users to get information about anything of his/her interest. But since the data is huge and complex, it is difficult to get useful information from its Recommender System are effective software techniques to overcome this problem. Based on the user’s and item’s information available, these techniques provide recommendations to users in their area of interest We are using two algorithm Euclidean distance and Cosine similarity to recommend books. In future, this work can be extended to recommend books with higher performance by adding big data tools. People find it difficult searching for books based on their preferences or choices were searching is done through various sites on the internet and finally determines which book is appropriate for buying or reading or referring to, which is a very tedious job. In this project we use Artificial intelligence to provide easy access to all the people who are in search of different varieties of books by providing them with recommendations, based on their reviews, likes, preferences using hybrid-based filtering techniques. It is specially designed to collect, record, store, count and display results accurately. This machine allows users to get the best book recommendation based on their preference with high accuracy.

Keywords: Personalize, Book Recommendation, Euclidean, cosine similarity

Abstract: Sign language is a main mode of communication for vocally disabled. This language uses a variety of symbols, including finger signs, expressions, and a combination of the two, to express information. This system offers a novel method for translating sign action analysis, recognition, and generation of a written description in Kannada language using mobile applications. Training and testing are two crucial procedures that are used. Each domain in the training set has 5 video samples, and each video sample has a class of words associated with it that will be stored in a database. Pre processing on the test sample is done using the median filter, the clever operator for edge detection, and the HOG for feature extraction.The text description will finally be produced in English. The performance efficiency over the traditional model is validated by the minimal average computation time, acceptable recognition rate, and performance.

Abstract: As the use of the internet is growing exponentially, more and more businesses such as the financial sector are initiating their services online. Accordingly, financial fraud is increasing in number and forms around the world, which results in financial losses which make financial fraud a major problem. Unauthorized access and immense regular attacks are examples of threats that should be detected by means of financial fraud detection systems. Machine learning and data mining techniques have been extensively used to tackle this problem over the past few years. However, these methods still need to be improved in terms of fast computation, dealing with huge data, and identifying the unknown attack patterns. Therefore, in this paper, deep learning-based method is implemented for the detection of financial fraudulence based on the Long Short-Term Memory (LSTM) technique and Bidirectional encoder representation from transformers (BERT) . This model is aimed at enhancing the present detection techniques as well as enhancing the detection accuracy in the light of big data. To evaluate the proposed model, a real dataset of credit card frauds is utilized and the results are compared with an existing deep learning model named spiking neural network and some other machine learning techniques. The experimental results illustrated a perfect performance of LSTM where it achieved 99.95% of accuracy.

Keywords: Fraud ; fraud detection; deep learning; long short-term memory

Abstract: Skin cancer affects 30,000 people each year, according to the World Cancer Research Fund. Most frequently, skin exposed to the sun gets skin cancer, an abnormal development of skin cells. But this typical sort of cancer can also develop on parts of your skin that aren't usually exposed to sunlight. Melanoma and Benign are the two main kinds of skin cancer. It is currently very difficult to automatically diagnose different skin lesion disorders using medical dermoscopy images. In this study, a cascading innovative deep learning network-based integrated model for segmenting skin lesion boundaries and classifying skin lesions is proposed. In the first stage, the boundaries of skin lesions are segmented from dermoscopy pictures using a unique full resolution convolutional network (FrCN). Following that, a deep residual network is fed with the segmented lesions to classify them.

Keywords: Skin cancer; Deep learning ; Dermoscopy ; Full resolution convolution network

Abstract: Internet of Things (IoT) is an emerging Technology with lots of opportunities in future endeavors. It can develop ease on different task for us but on the other hand it causes security threats like data breaches, Data authentication and virus. Some classic cryptography algorithm like RSA (Rivest-Shamir-Adleman) works under the classical computers. But the newly acquired Technology is shifting towards Quantum cryptography easily. Therefore it is much required to design Quantum cryptography algorithm to prevent our system from security breaches. IoT will also be one of the discipline which needs to be secured to prevent any malicious activities. In this paper we are going to review the common security problems in IoT and there presently available solutions with their drawbacks. Then the analysis has been carried out in terms of the advantages and disadvantages of implementing Quantum cryptography for the IoT security. And finally Quantum cryptography is introduced with some of its variations.

Keywords: Quantum cryptography, Quantum computing, cryptography, security, internet of things.

Abstract: Long videos captured by consumers are typically tied to some of the most important moments of their lives, yet ironically are often the least frequently watched. The time required to initially retrieve and watch sections can be daunting. In this work we propose novel techniques for summarizing and annotating long videos. Existing video summarization techniques focus exclusively on identifying keyframes and subshots, however evaluating these summarized videos is a challenging task. Our work proposes methods to generate visual summaries of long videos, and in addition proposes techniques to annotate and generate textual summaries of the videos using recurrent networks. Interesting segments of long video are extracted based on image quality as well as cinematographic and consumer preference. Key frames from the most impactful segments are converted to textual annotations using sequential encoding and decoding deep learning models. Summarization technique is benchmarked on the VideoSet dataset, and evaluated by humans for informative and linguistic content. We believe this to be the first fully automatic method capable of simultaneous visual and textual summarization of long

Keywords: Natural Language Processing, Deep Learning, Convolutional Neural Networks, Sequence to Sequence, Word2vector and Deep Speech Package.

Abstract: The image speaks a thousand words as the advancement in the image processing technology combined with mini cameras and cheap storage made every one photographer as the images are stored locally and in the cloud. The growth of the social media made every one connected by this user post are shared often in the form of an image. With the growth of the image processing technology, the growth of the image manipulation and improvement software has been on the raise as lot of the software is mobile device based. As is the case with other technologies the image manipulation software has its own drawbacks. By using the software the personal identity can be changed to the bad affect this causes the loss of personal identity also results into identity assassination of an individual. Image manipulation software also manipulate the situations in an image which is the primary source of the fake news. The current system also attempts to solve the current problem by detecting the manipulation in an image. The system uses CNN architecture to point out the area of which the image is manipulated and displayed in the window.

Abstract: News and Media has become one of the most significant components of human existence, based entirely on the most recent generation and traits within the discipline of laptop generations. This area has become a famous platform for sharing information and statistics on a variety of issues, as well as daily reports, and is the most popular generation for transferring and generating data. There are various advantages to living in this environment. however there also are several fake statistics and data that mislead the reader and consumer with a view to acquire the numbers required. One of the system's primary trouble is the shortage of functional data and real statistics on social media data. To resolve this issue, we have got proposed an incorporated system with Various additives of the Convolutional Neural Network (CNN) and Natural Language Processing (NLP) to utilize machine learning analyzing strategies to discover faux statistics and better are anticipating faux money owed and posts. This approach is finished the use of the Reinforcement Learning method. To enhance the safety of this platform, the decentralized blockchain framework became implemented, which incorporates the definition of digital content material authority proofs. More specifically, the purpose of this system is to expand a dependable platform for awaiting and detecting social media networks.

Keywords: CNN (Convolutional neural network) Algorithm,NLP (Natural Language Processing)

Abstract: In recent years, due to the COVID-19 pandemic, Search Engine Optimization (SEO) has hit an all-time high. As consumers shifted to online shopping in droves, even the most traditional businesses realized they needed to catch up with the digital shift. SEO is important since it enhances the visibility of websites, which provides more traffic and possibilities to convert leads into consumers. SEO may help you figure out what people are looking for on the internet, what answers they're looking for, what terms they're using, and what kind of content they want to consume. The explanations will enable site owners to reconnect with people who are looking for solutions to their problems online. Because most internet users don't look above the first page of search results, the higher a website ranks for a specific term or phrase, the more likely it is to attract new clients. In recent years, practically all website owners have been scrambling to boost their search engine rankings. Business digitization has become a requirement. The situation hasn't changed once the website is built you'll need to implement a variety of digital marketing methods to boost your business's visibility. The purpose of our project Search engine optimizer (SEO) suggester is to help achieve the same. The solution is to build a system that will take as input a websites URL and the users can choose to audit the website or compare the website with the higher ranked website in the domain. The system then produces the summary of what changes can be done in terms of keywords used on the website and changes that can be done in the structure of website and report it to the user.

Keywords: Search Engine Optimization (SEO), Website rank, Search Engine result page (SERP), Website

Abstract: Disease detection in plants plays a very paramount role in agriculture. Disease in plants causes major endangerment and economic losses in agriculture industry ecumenical. Prognostication of crop health and disease early can facilitate the control of diseases. Magnification of plant is major requisite of farmers as they are a paramount aspects of ones survival, as the pabulum demand is incrementing at an expeditious rate due to an incrementalism in population. Moreover, the utilization of technology today has incremented the efficiency and precision of detecting diseases in plants. These techniques are applied to detect diseases from infected plants. Getting affected by a disease is very prevalent in plants due to sundry factors such as fertilizers, cultural practices followed, environmental conditions, etc. These diseases hurt agricultural yield and ineluctably the economy predicated on it. Plant disease detection utilizing image processing is the best way to detect and get exact results. This application will avail farmers to ken the correct information of the disease and avail in increase their yield. The moto is to detect sundry plants diseases and provide precautions and remedies to preserve the plants from eradicating.

Keywords: Plant Diseases, Machine Learning, Image Processing, CNN, Plant Village.

Abstract: Recommender systems are software tools used to leverage different strategies to generate suggestions for movies and other entities and make them available to users. Hybrid recommender systems combine two or more recommendation systems in different ways to take advantage of their complementary benefits. This systematic literature review reveals the latest technology in hybrid recommender systems over the last decade. An overview of relevant data mining and recommended techniques used to address and overcome the most relevant issues under consideration. It also considers the hybridization class to which each hybrid recommender belongs, the application domain, the evaluation process, and the proposed future research direction. Based on our results, most studies combine collaborative filtering with other techniques and are often weighted. The Hybrid Recommender System is a hot topic and provides a good foundation for responding to new opportunities by exploring new opportunities, Contextualization recommendations, embedding parallel hybrid algorithms, and processing large datasets.

Keywords: Movie Recommender System, Hybrid Recommender System, Content – Based System, Collaborative based System.

Abstract: With the advent of statistical modelling in sports, predicting the outcome of a game has been established as a fundamental problem. Game consists of 11 player team sport played on ground. Cricket has huge fan base in India. With are great spectator support and many people try to predict the outcome of matches based on their individual cricket sense. The games has some rules and scoring system. Factors viz, match location and individual player performance have great impact on outcome of the match. Such various parameters are highly interdependent on each other which makes it heard to make precise prediction of the match. In this project, we are going to build prediction system that takes in data of matches played in past and makes a prediction of future match events such as final score and results in a gain or loss. Our system will predict match outcome by analysing pre-stored match data using various machine learning algorithms . We intend to use more features such as pitch condition, weather condition, outcome of toss, individual player performance with respect to match venue. Our system finally present quantitative results displayed by best suited algorithm having highest accuracy. Also, demonstrating the performance of our algorithms in predicting the number of runs scored which is one of the most important parameter of match outcome This work suggests that the relative team strength between the competing teams forms distinctive feature for predicting the winner. Modeling the team strength boils down to modeling individual player's batting and bowling performances, forming the basis of proposed approach. The career statistics as well as the recent performances of a player to model this. Player independent factors have also been considered in order to predict the outcome of match . The algorithm used are Decision Tree, Logistic regression and Support Vector Classifier (SVC) yields better results in the experimental evaluations.

Keywords: Prediction ; Cricket Match outcome prediction; Machine learning techniques

Abstract: Crypto-currency such as Bitcoin is more popular these days among investors. In the proposed work, it is attempted to predict the Bitcoin price accurately taking into consideration various parameters that affect the Bitcoin value. For the first phase of investigation, it is aimed to understand and identify daily trends in the Bitcoin market while gaining insight into optimal features surrounding Bitcoin price. The data set consists of various features relating to the Bitcoin price and payment network over the course of time, recorded daily. For the second phase of investigation, using the available information, we will predict the sign of the daily price change with highest possible accuracy

Abstract: Robotic Process Automation (RPA) has received growing attention within the digital transformation. However, the adoption in the educational sector is hardly been explored. The traditional method of result analysis has certain drawbacks. Checking a large number of student results and storing the data in an excel file will be very difficult. It is not time efficient as the process is repeated for each student. There is no automated technology that is implemented for such a process. In this project, we are using RPA technology which automates the process and is very useful for the educational institute. RPA is used to get rid of human tasks, every automation is supposed to achieve something, so knowing and defining that end goal is important. It is specially designed to collect, record, store, and display results. Instead of recruiting and training new employees, software bots support and replace human operators by automating their behaviour.

Keywords: Robotic Process Automation, UiPath.

Abstract: One of the risks of getting a medical diagnosis is getting breast cancer, and the number of people with it is growing all over the world. This causes the female to die day by day. The main goal of the work is to get the ROI and stages of the Breast Cancer Mammogram images. This work consists of three modules, i.e., preprocessing, segmentation and feature extraction and classification. Image is preprocessed by removing noise element and converting RGB to gray scale image. By applying morphological operation to obtain ROI for the tumor. This tumor is classified to find the stages of the tumor by multi SVM classifier.

Keywords: - Breast Cancer, Mammogram images, ROI, multi SVM classifier.

Keywords: home and network intrusion detection system,HNIDS,ML,GAN,LSTM

Abstract: In India, ovarian cancer is the third most frequent malignancy. Every year, it affects over a lakh people. In 2018, 295,414 women were diagnosed with ovarian cancer, and 184,799 women died from the disease globally, according to statistics. in their life. At some time in their life, one out of every 78 women will develop ovarian cancer. Because early-stage tumors are often asymptomatic, the vast majority of ovarian cancer patients are diagnosed with advanced disease. As a result, long-term survival appears to be improbable. To find out if you have this cancer, consult a doctor or travel to a diagnostic center, which can take time. A few statistical-based approaches to dealing with this problem are currently being explored, and they have become part of a partial answer to some extent. In the healthcare industry, machine learning has made a wide range of tools, methodologies, and frameworks available. Machine learning is the most effective connectionist technique for predicting cancer outcomes because it can identify and recognize patterns in complex datasets. Intending to lower mortality rates, this project provides a set of classification-based machine learning algorithms for cancer detection and prevention. Our goal is to create a simple predictive model that also performs well. It's done using classification techniques including Decision Tree (DT), Logistic Regression (LR), and Support Vector Machine (SVM) (SVM). The accuracy of each categorization method's conclusions is compared.

Keywords: Ovarian cancer, Decision Tree (DT), Logistic Regression (LR), Support Vector Machine (SVM).

Abstract: The custom music recommender supports users' favorite songs, which are stored in a huge music database. To predict only the user's favorite songs, the management of the user's preference information and the genre rating is required. In our study, a very short feature vector obtained from a low-dimensional projection and already developed audio features is used for the music genre classification problem. We apply a metric distance learning algorithm to reduce the dimensionality of the feature vector with little performance degradation. We propose the system through the automatic management of user preferences and gender classification in the personalized music system. This Recommender System uses a feature level fusion to combine multiple perspectives and gives an outcome that suits all types of users. The performance of this system is compared with existing legacy system.

Abstract: As we can see in today's generation the hospital records and data are stored more in the cloud. As well as the data of the patient needs to saved more securely. The security of the data and the information security is the basic important in most of the organization , even in the home computer user, client data, data of payment, documents and account details. This data are difficult to change, or it may be difficult to replace, and it becomes danger, if it goes to the wrong hand. The gathered information which are lost due to the flood, fire is a type of crushing but loosing of data by the hacker can have significantly more net worth result. In other word we can say that when the block is completed , It creates a secure code ,In this system the data will be of the health care system, and it needs to secure this work is being designed by the help of blockchain concept key based cryptographic technique it works on the storing data of the health care, and it is stored on the web that stored data of the health care should be in the secure mode.

Keywords: Block-chain technology, ledger, cryptography, hashing, healthcare system storage data.

Abstract: Power BI has taken the world of business intelligence, data visualization and analytics by storm. Power BI is an online service that enables searching data, transforming it, visualizing it, and sharing the developed reports and dashboards with other users in the same or different department/organizations or even with the general public. As of February 2017, more than 200,000 organizations across 205 countries are using Power BI. Power BI is having a free option that has adequate features and functionality, One of the innovative features of Power BI is its Quick Insights feature (Michael Hart, 2017) that is built on a growing set of advanced analytical algorithms. The above paragraph describes our project by the built dashboard which is made up of data that is collected from “Julco” website, which is full of Cryptocurrency data analysis, by this someone can predict the amount of cryptocurrency been sale by the bitcoins and also it can be shown in the various strategies like how much dollar been spend to buy the bitcoins.

Abstract: In this research, a system for accurately identifying bird species was developed, and tactics for identifying them were studied. Automatic recognizing bird sounds with no physical interaction has proven to be a challenging and time-consuming task for significant research in ornithology's taxonomy and other subfields. Birds make a wide range of vocalisations, and different species of birds have distinct functions. Manual annotation of each recording is used in current methods for processing big bioacoustic datasets. This necessitates specialised expertise and an inordinate amount of time. Recent advances in machine learning had made it easier to identify specific bird sounds for popular species with enough training data. However, developing such tools for rare and endangered species remains difficult. This problem has been addressed in two stages : pre-processing and modelling (CNN model). The first step is to create a spectrogram from an audio input. The spectrograms were used as input in the second stage, which required establishing a neural network. Depending on the input properties, the Convolutional Neural Network identifies the sounds clip as well as separates the bird species. Key-words: Bird Species Classification , Bird audio , method of pre-processing of bird sound, Convolutional Neural Network (CNN), Spectrogram

Abstract: Psychological issues among students, including as depression and anxiety, are mostly caused by a failure to check students' psychological well-being on a regular basis. If depression is discovered early enough, it can be treated successfully. Image processing advancements have resulted in the development of successful algorithms that can identify emotions from facial pictures in a much more basic manner. As a result, for successful diagnosis of depression, we require an automated system that records and analyses student facial photos. In the proposed system, image processing techniques are used to analyse student facial features and depression can be predicted. To predict depression, a video of the student is gathered, and the student's face is identified by use of haarcascade classifier. The Mini Xception classifier is used to classify these facial traits. After quantifying the system, we will be able to implant it in mobile devices in the future.

Abstract: Millions of users throughout the world are active on social networking sites. Users' interactions with social media platforms like Twitter and Facebook have a significant impact on daily life, sometimes in unfavourable ways. Popular social networking sites have become a target for spammers who want to spread a tonne of harmful and unnecessary content. Twitter, for instance, has grown to be one of the most extravagantly used platforms ever and as a result, permits an excessive quantity of spam. False users spam users with unwanted tweets to advertise products or websites that not only negatively impact real users but also disturb resource usage. A popular field of research in today's online social networks is the identification of false Twitter users and the detection of spammers (OSNs). Review the procedures for identifying spammers on Twitter. In addition, a taxonomy of Twitter spam detection methodologies is offered, which groups the methods according to how well they can identify I phoney material, (ii) spam based on URL, (iii) spam in trending topics, and (iv) fake users. The presented techniques are also contrasted based on a number of criteria, including user, content, graph, structure, and time factors. We are optimistic that the study that has been provided will serve as a beneficial tool for scholars looking for the most significant recent advancements in Twitter spam detection on a single platform.

Keywords: Spam Detection, Fake user, online social networks, Detecting URL

Abstract: Research of quantitate investment on stock price prediction is effective to help investors increase profits. Recently, technologies of machine learning have been well applied to explore the issue of stock trading. In this paper, Logistic Regression and Support Vector Machines (SVM) were adopted to solve the problem of predicting the trend of stock movements. The experiment showed that these two models could be effectively used in the stock market of China. Returns based on strategies we constructed were significantly better than the HS300 index. We investigated the relationship between stock returns and various models using various models. It found that the SVM model results are optimal. The annual return of the strategy based on SVM reached 17.13% and the maximum Drawdown was 0.32. In the future, we will not only focus on the stock market, but also plan to apply this strategy to other investment fields, such as trading of digital currency. We will also use other algorithms for research and comparison

Abstract: Cloud computing may be a general term that involves delivering hosted services over the world wide web. These services are divided into three main categories: infrastructure as a service (IaaS), platform as a service (PaaS) and software as a service (SaaS). Cloud infrastructure involves hardware and software components for proper implementation of a model. Cloud computing can be thought of as utility computing or on-demand computing. Cloud computing works by enabling client devices to access data and cloud applications over the web from remote physical servers, databases and computers. With the widespread use of cloud emails and frequent reports on email leakage events, a security property called forward secrecy becomes desirable for both individuals and cloud email service providers to strengthen the protection of cloud email systems. Specifically, forward secrecy can guarantee the confidentiality of these previously encrypted emails whether or not the user’s secret key gets exposed. However, thanks to the failure to satisfy the protection requirements of email systems simultaneously, typical methods like Diffie-Hellman key exchange and forward-secure public-key encryption, haven't been widely approved and adopted. To capture forward secrecy of encrypted cloud email systems, we introduced a replacement cryptographic primitive called forward-secure puncture able identity-based encryption (fs-PIBE), which enables the user to perform fine-grained revocation of decryption capacity. in additional detail, the user is allowed to preserve the decryption capacity of unreceived encrypted emails while abolishing that of these received ones. Thus, it provides more practical forward secrecy than typical manners, during which the decryption capacity of received and unreceived encrypted emails is revoked simultaneously. supported such a primitive, we build a framework of encrypted cloud email systems.

Keywords: cloud emails, Forward secrecy, Diffie-Hellman, forward-secure public-key encryption, identity-based encryption, fs-PIBE.

Abstract: By increasing human health, safeguarding aquatic habitats, and lowering air pollution, smart cities do have the ability to tackle the environmental issues provided by appropriate waste management. In this article, we look at how blockchain technology might help smart cities manage waste by allowing for traceability, immutability, transparency, and audits in a decentralised manner, dependable, and secure way. We discuss the advantages of blockchain technology in a variety of waste management situations, such as real-time garbage cans monitoring and tracking, reliable waste channelling and compliance with waste management rules, efficient waste resource management, waste management data preservation, and material handling are all examples of waste management services. To focus to the usefulness of blockchain technology in smart city waste management, we report on many projects and case studies based on blockchain technology. We identify and address a number of open research challenges that are impeding the successful deployment of blockchain in management of waste in smart cities.

Keywords: Waste Management, Blockchain, Decentralized, Transparent.

Abstract: The development of cloud computing era with the explosive increase of unstructured records, cloud storage era gets more interest and better development. The cloud provider does not have pointers regarding the statistics and the cloud facts saved and maintained globally everywhere within the cloud. The privateness safety schemes are typically based totally mostly on encryption era. there are many privateness keeping strategies in the facet to prevent facts in cloud. We propose a three factors of storage framework primarily based on fog computing. The proposed framework can each take whole benefit of cloud storage and guard the privacy of data. here we're the use of Hash-Solomon code set of rules is designed to divide statistics into specific components. If the simplest facts aspect missing we lost the information records. on this framework we are using bucket idea based totally algorithms and secure the records data after which it could show the security and performance in our scheme. moreover, based on computational intelligence, this set of regulations can compute the distribution percent stored in cloud, fog, and local system. patron releases their utility on a hosting environment which can be accessed through network from diverse clients via application users.

Keywords: cloud Computing, Pointers, Hash-Solomon, privateness, Encryption, Fog Computing, Computational intelligence

Abstract: The fake news on social media and various other media is wide spreading and is a matter of serious concern due to its ability to cause a lot of social and national damage with destructive impacts. A lot of research is already focused on detecting it. This paper makes an analysis of the research related to fake news detection and explores the traditional machine learning models to choose the best, in order to create a model of a product with supervised machine learning algorithm, that can classify fake news as true or false, by using tools like python scikit-learn, NLP for textual analysis. This process will result in feature extraction and vectorization; we propose using Python scikit-learn library to perform tokenization and feature extraction of text data, because this library contains useful tools like Count Vectorizer and Tfidf Vectorizer. Then, we will perform feature selection methods, to experiment and choose the best fit features to obtain the highest precision, according to confusion matrix results.

Keywords: Fake news, TF-IDF,SVM,feature extraction,training classifier.

Abstract: Machine studying has large programs in the finance industry. danger Analytics, consumer Analytics, and inventory marketplace Predictions are a number of the domains where gadget mastering techniques can be implemented. accurate prediction of inventory marketplace returns is extremely tough because of volatility inside the marketplace. the main factor in predicting a inventory market is a excessive stage of accuracy and precision. With the creation of artificial intelligence and excessive computational potential, efficiency has increased. in the past few a long time, the surprisingly theoretical and speculative nature of the inventory market has been tested by capturing and the use of repetitive patterns. numerous gadget mastering algorithms like more than one Linear Regression, ARIMA model, Random forest set of rules, and so forth. are used here. The financial records incorporates elements like Date, volume, Open, high, Low close, and near prices. The fashions are evaluated the use of fashionable strategic signs LSTM and R2 score. lower values of those two indicators mean better efficiency of the educated fashions. various agencies hire distinct types of analysis gear for forecasting and the primary goal is the accuracy to achieve the most earnings. The a success prediction of the stock may be a useful asset for the inventory marketplace institutions and could provide real-existence solutions to the issues of the traders.

Abstract: Handwritten recognition is the ability of a computer to acquire and interpret handwritten entries from sources including paper files, images, touch displays, and different gadgets. Handwritten reputation is the maximum difficult undertaking because it's far a repeated painting that's written through human beings and inflicts terror. The handwritten reputation has been carried out in form of programs like Banking sectors, Health care industries, and lots of such organizations. Handwriting reputation is maximum generally utilized in modern-day cellular international is handwriting reputation as a right away enter to be transformed into texts. This is beneficial because it lets the person to speedy jot down numbers and names for contacts compared to inputting the equal records thru the onscreen keyboard. This paper consists of a survey on handwritten textual content reputation studies papers that have used IAM datasets.

Keywords: Handwritten recognition, IAM dataset.

Abstract: Despite everything that has been done to improve road safety in India to date, there are always problems lurking around every corner. This circumstance has shown a problem with traffic accidents, affecting public health and the economy of the country. In the past, it was assumed that road accidents and fatalities could not be avoided, but in today's technological age, anything is almost possible. Our research aims to reduce mortality rates by developing a prediction model that considers factors like carriageway/roadway hazards, light conditions, day of the week, special conditions at the accident scene, road class, junction control, junction details, road surface condition, road type, and weather conditions. Machine learning techniques such as random forests (RF), logistic regression (LR), and naive Bayes (NB) were used to create these models. The goal of this research is to analyze data on road accidents in India utilizing the best compatible machine learning classification approaches for estimating road accidents through data mining. Our findings suggest that logistic regression outperformed other machine learning algorithms in terms of accuracy.

Abstract: Traffic congestion not only. Affecting the human being, but also elevates the pollution .The most important causes that increase traffic congestion are lack of planning of city road, using vehicle widely .Traffic monitering has only the manual algorithm that is the man power. In manual use we can only able to analysis the vehicles in one particular direction, so the accident or vehicles that passing through the another road or side can’t be captured .Vehicle speed estimation used to calculate the speed in each image frame by using the vehicle position in each image frame .In this process it is possible to widely apply deep learning method to the analysis of traffic surveillance video. Traffic flow prediction ,anomaly detection, vehicles re-identification and vehicle tracking are basic components in traffic analysis. Among the application traffic flow prediction or vehicle speed estimation is one of the most important research topics of recent years .This project proposed the convolution neural network algorithm and hard clustering method will used to calculate the speed estimation. By using this process it will collect the details of the vehicles types, speed of the vehicle and vehicle detection.

Keywords: Machine Learning, Traffic monitering, Anomaly detection

Abstract: Agriculture is a major source of income in India, and the country's economy is heavily reliant on it. To maximize agricultural productivity and profit, it is critical to diagnose plant leaf diseases at an early stage. Because naked eye observation of diseases does not always yield reliable results, especially during the early stages, an image processing technique is utilized to detect leaf diseases accurately. It consisted of five steps: image acquisition, preprocessing of the acquired image, feature extraction, disease classification, and display of the results.. This work presents a thorough examination of the categorization of agricultural illnesses using the Support Vector Machine classifier. Recently, many works have been inspired by the success of deep learning in computer vision for plant diseases classification. Unfortunately, these end-to-end deep classifier slack transparency which can limit their adoption in practice. In this paper, we propose a new trainable visualization method for plant diseases classification based on a Convolutional Neural Network (CNN) architecture composed of two deep classifiers. The first one is named Teacher and the second one Student. This architecture leverages the multitask learning to train the Teacher and the Student jointly.Then, the communicated representation between the Teacher and the Student is used as a proxy to visualize the most important image regions for classification. This new architecture produces sharper visualization than the existing methods in plant diseases context.

Keywords: Image processing , Save Model, CNN, Leaf Disease, Matlab, Feature Extraction, Deep Learning, Leaf Dataset.

Abstract: Lately, the matter of dangerous malware in devices is spreading speedily, particularly those repackaged android malware. Though understanding robot malware mistreatment dynamic analysis will give a comprehensive read, it's still subjected to high price in setting preparation and manual efforts within the investigation. Android is the most preferred openly available smart phone OS and its permission declaration access management mechanisms can’t sight the behavior of malware. the matter of police investigation such malware presents distinctive challenges thanks to the restricted resources accessible and restricted privileges granted to the user however conjointly presents distinctive opportunities within the needed data hooked up to every application. In our project, a code behavior signature-based malware detection framework mistreatment associate degree SVM rule is planned, which might sight malicious code and their variants effectively in runtime and extend malware characteristics information dynamically. Experimental results show that the approach incorporates a high detection rate and low rate of false positive and false negative, the power, and performance impact on the first system can even be unheeded. Our system extracts variety of options associate degreed trains a Support Vector Machine in an offline (off-device) manner, so as to leverage the upper computing power of a server or cluster of servers.

Keywords: Support Vector Machine,Svm Classifier,Malware

Abstract: Chronic Kidney Disease (CKD) is one of the deadliest diseases that slowly damages human kidney. The disease remains undetected in its early stage and the patients can only realize the severity of the disease when it gets advanced. Hence, detecting such disease at earlier stage is a key challenge now. Machine Learning is one of the emerging field used in the health sectors for the diagnosis of different diseases. In this paper, we compute, analyze and compare between Machine Learning classification approaches to determine which classification approach is the optimal for the prediction of CKD. Random Forest Algorithm and Logistic Regression are some renowned machine learning methods which were selected to train the model and based on these results, we can compare and determine which among the following Machine Learning Methods can predict the possibility of CKD at the most accurate level. From this comparative analysis, Random Forest Algorithm is found to be the best approach to predict CKD. Methods can predict the possibility of CKD at the most accurate level. From this comparative analysis, Random Forest Algorithm is found to be the best approach to predict CKD.

Keywords: Machine Learning, Classification Technique, Prediction System

Abstract: Cybercrime has become more prevalent in recent years. With modern photo editing tools as widely available as ever, it has been demonstrated that creating phony papers is incredibly simple [1]. With the help of this tool, which offers tools for doing so, documents can be scanned and forged in minutes. While photo editing software is convenient and widely available, there are also deft methods for investigating these transformed documents. This study presents a framework for investigating digitally modified documents as well as a way of distinguishing between an original document and a digitally morphing document. we created a web application to detect digitally modified photos. This method has more than 95.0 percent accuracy and has proven to be efficient and useful. Recent work on forgery detection using neural networks has proven to be very effective in detecting image forgery additionally we are using Azure Form recognizer service to read data from documents and verify it on the server, this dual approach makes the system robust and very accurate. Deep Learning methods are capable of extracting complex features in an image, resulting in increased accuracy. In contrast to traditional methods of forgery detection, a deep learning model automatically builds the required features, and as a result, it has emerged as a new area of study in image forgery.

Keywords: morphed document, Azure form recognizer, CNN, Deep learning, neural network.

Abstract: This project presents a method for identifying apple disease and an approach for careful detection of diseases. The goal of proposed work is to diagnose the disease of fruit using image processing and (SVM) Support Vector Machine. The diseases on the fruit are critical issue which makes the sharp decrease in the production of fruit. The study of interest is the fruit rather than whole apple fruit fruit because about 85-95 % of diseases occurred on the fruit like, the methodology to detect fruit disease in this work includes K mean’s clustering algorithm for segmentation and Support Vector Machine. The proposed detection model based very effective in recognizing fruit diseases.

Keywords: Image Processing,Support Vector Machine,K means Clustering

Abstract: The challenge entitled health practitioner Appointment machine the use of React local turned into designed and evolved in React local, MySQL. The motive of the challenge is to put in force a web database gadget that caters the doctor appointment scheduling, reservation and records control of consultation for a sanatorium or medical institution. In this period of pandemic, clinical clinics and hospitals imposed strict recommendations at the variety of those who can enter their facility. Consultations to medical doctors are also very limited and might facilitate a constrained number of patients. the net machine for appointment and consultation is one of the answers that can be used with the intention to give the humans a manner on the way to touch their medical doctors and reserve an appointment for consultation. The implementation of the said undertaking will assist docs offer better purchaser control while ease and luxury is the gain on the affected person cease. The evaluation and implementation result suggests that the challenge is feasible and implementation is exceptionally recommended by way of the team.

Keywords: React,MySQL,Node JS

Abstract: Emotions describe the physiological states of an individual and are generated subconsciously. They motivate, organize, and guide perception, thought, and action. Emotions can be positive or negative. Negative emotions manifest in the form of depression, anxiety and stress. It is necessary to identify negative emotions of an individual who might be in the need for counselling or psychological treatment. Body signal analysis, handwriting analysis, and psychological assessment are some mechanisms to measure them. It’s a key factor for every individual to keep his mental health well. Among the various reasons for causing illness, some of them can be lot of tension, stress and depression. It is important to know if a person is suffering from depression or stress, which are negative emotions. The positive emotions are joy, happiness, excitement etc. while negative emotions are sadness, anger, disgust, fear, depression, anxiety and stress.

Abstract: Spying literally way to watch from a distance. in this task, the paintings of designing and implementing manner of an Android primarily based secret agent robotic is described .the primary reason in the back of growing this robot is for the surveillance of human sports inside the conflict subject or get to realize about the state of affairs if a disaster happens, to decrease the risk of a lifestyles .The robot along with camera can wirelessly transmit actual time video with night time vision abilties, and with multi-sensor capabilities .This robotic is used to collect information from the far off terrain and reveal that information at a miles cozy vicinity with the help of IOT era. it's far widely used due to its simplicity and capability to regulate to fulfill alternate of wishes. The video transmission is practically achieved via high-speed photo transmissions. The robot will serve an appropriate system for the defense quarter and will also prevent unlawful activities.The robotic will help the police or navy/military to recognize the situation of the territory earlier than getting into it.

Keywords: Arduino, Spying Robot, LCD sensor, Montor, WiFi Control.

Abstract: Security is at most concern for anyone nowadays, whether it's data security or security of their own home. With the advancement of technology and the increasing use of IoT and AI, digital door locks have become very common these days. Face recognition system is broadly used for human identification because of its capacity to measure the facial points and recognize the identity in an unobtrusive way. The application of face recognition systems can be applied to surveillance at home, workplaces, and campuses, accordingly. The problem with existing face recognition systems is that they either rely on the facial key points and landmarks or the face embeddings from FaceNet for the recognition process. Deep convolutional neural networks have been successfully applied to face detection recently. Despite making remarkable progress, most of the existing detection methods only localize each face using a bounding box, which cannot segment each face from the background image simultaneously. To overcome this drawback, this project present a face detection and identification method based on improved Mask R-CNN, named G-Mask, which incorporates face detection and recognition into one framework aiming to obtain more fine-grained information of face. This paper also investigates the robustness of the face recognition system when an unknown person is being detected, wherein the system will send an SMS Web link to the owner of the house through edge computing. The door lock can also be accessed remotely from any part of the world by using a door lock integrated server account.

Keywords: IOT, AI, Deep Convolutional Neural Networks,Mask R-CNN

Abstract: Generally, predicting how the house price will perform is one of the most difficult things to do. It can be described as one of the most critical process to predict that. This is a very complex task and has uncertainties. To prevent this problem in One of the most interesting (or perhaps most profitable) time series data using machine learning techniques. Hence, house price prediction has become an important research area. The aim is to predict machine learning based techniques for house price prediction results in error based calculation. The analysis of dataset by supervised machine learning technique(SMLT) to capture several informations, missing value treatments and analyze the data validation, data cleaning/preparing and data visualization will be done on the entire given dataset. To propose a machine learning-based method to accurately predict the house price Index value by prediction results in the form of house price increase or stable state best regression from comparing supervise machine learning algorithms. Additionally, to compare and discuss the performance of various machine learning algorithms. dataset with evaluation classification report, to categorizing data from priority and the result shows that the effectiveness of the proposed machine learning algorithm technique can be compared with best accuracy MAE, MSE, R2.

Abstract: In training machine evaluation and prediction of pupil overall performance is a hard task. on this paper, a model is proposed to are expecting the overall performance of students in an academic agency. The set of rules hired is a system gaining knowledge of technique referred to as Naïve Bayes and KNN. similarly, the importance of numerous extraordinary attributes is considered, in an effort to decide which of those are correlated with student overall performance. finally, the results of an experiment observe, showcasing the electricity of system learning in such an software. In attitude of this mission we are going to expect the student development and look at the greater end result thru device learning set of rules. We foresee the scholar overall performance by scanning their preceding instructional details. To execute this prediction, we've created a dataset, via the usage of this we can predict pupil information. in many corporations, records mining techniques are used for studying huge amount of to be had facts, records for choice making method. In instructional quarter, statistics mining is used for wide form of utility such as performance of the scholars like mark, attendance, personnel opinion, extracurricular activities, Ragging and strain. The facts mining strategies used for figuring out the performance of the scholar the use of Naïve Bayes and KNN algorithms. those algorithms pick out and analyses the overall performance of the student.

Keywords: KNN, Naïve Bayes, Statistics mining, Prediction, Performance.

Abstract: The intention of this system is to reduce human victims in terrorist attack such as 26/11. So, this problem can be overcome by designing the RF based spy robot which involves wireless Devices, So that from this it will be easy to examine rivals when it required. This robot can quietly enter into enemy area and sends us the information via wireless Devices. Robotics has been a staple of advanced manufacturing for over half a century. As robots and their peripheral equipment become more sophisticated, reliable and miniaturized, these systems are increasingly being utilized for military and law enforcement purposes. Now-a-days android smart phones are the most popular gadget. There are multiple applications on the internet that exploit inbuilt hardware in these mobile phones, such as Bluetooth, Wi-Fi and ZigBee application control. Here we have designed a robot that can be controlled using an application running on an android phone. It sends control command via Bluetooth which is interfaced to the controller. The controller can be interfaced to the Bluetooth module though UART protocol. According to commands received from android the robot motion can be controlled. And hence the required actions can be taken. But object tracking is one of the major fundamental challenging problems in computer vision applications. This project presents a helpful application with a real-time object detection system that can automatically capture the user-defined important objects.

Keywords: Bluetooth module HC-05, Metal Sensor for metal detection, Temperature Sensor for temperature Detection, Ultrasonic Sensor for obstacle detection, Data transparency.

Abstract: Cloud computing is a platform for dynamically growing capabilities and increasing potentialities without the need for new infrastructure, staff, or software. Furthermore, cloud computing began as a commercial enterprise notion and has since grown into a thriving IT technology. However, considering that the cloud stores a great deal of information about individuals and businesses, worries about the cloud's security have been raised. Despite all of the buzz around cloud computing, clients are still hesitant to move their businesses to the cloud. Nonetheless, the lack of security is the single major problem that is preventing more people from using cloud computing. Furthermore, the market is wary about cloud computing due to the intricacy with which it controls data confidentiality and information security.When established technologies are employed in a cloud environment, the architecture of cloud models poses a security risk. As a result, users of cloud services should be aware of the risks associated with uploading data into this new environment. As a result, numerous cryptographic features that represent a threat to cloud computing are examined in this research. This study examines the security concerns raised by the usage of cryptography in a cloud computing environment. Index Terms: Cloud encryption, cryptographic algorithms, cloud security infrastructure.

Abstract: massive records is widely considered as potentially the following dominant technology in IT enterprise. It gives simplified machine renovation and scalable aid control with garage structures. As a essential era of cloud computing, storage has been a warm studies subject matter in current years. The high overhead of virtualization has been properly addressed through hardware development in CPU enterprise, and by means of software program implementation improvement in hypervisors. however, the high demand on garage picture garage stays a challenging hassle. existing systems have made efforts to lessen garage picture storage consumption via deduplication inside a storage location network system. nonetheless, storage place community can't satisfy the increasing demand of big-scale garage web hosting for cloud computing due to its value quandary. on this challenge, we suggest SILO, a scalable deduplication report device that has been in particular designed for huge- scale garage deployment. Its design presents rapid storage deployment with similarity and locality based totally fingerprint index for records transfer and low garage consumption by way of deduplication on storage photos. It additionally affords a complete set of storage capabilities including on the spot cloning for storage pics, on-call for fetching through a network, and caching with local disks by way of copy-on- examine strategies. Experiments display that SILO functions perform properly and introduce minor overall performance overhead.

Keywords: Deduplication ,Storage ,Security ,SILO function.

Abstract: On Android, every application operates during a basic sandbox and is prevented from accessing additional services that need users' consent. These services can only be accessed if users allow the appliance to use them. Granting of permission is static and may only be done at the time of the installation of the application. Android security model leaves most of the labor for security-related approaches to the user. LibreAV is an endeavor to detect malware on Android devices employing a machine learning approach that is powered by the Tensor flow. We use a two-layer neural network trained with a set of features. The neural network is tuned in such the simplest way that it performs efficiently on mobile devices where computational resources are limited. Testes show that LibreAV performs efficiently and effectively even on low-end mobile devices. With LibreAV, you’ll be able to scan all the installed apps on your device in an exceedingly matter of seconds. It also encompasses a real-time scan feature that alerts you whenever an app is installed or updated. Neural networks and SHA scan the applications and predict potential malicious behavior using state-of-the-art Machine Learning algorithms.

Abstract: A credit card is issued by a bank or financial services company that allows cardholders to borrow funds with which to pay for goods and services with merchants that accept cards for payment. Nowadays as everything is made cyber so there is a chance of misuse of cards and the account holder can lose the money so it is vital that credit card companies are able to identify fraudulent credit card transactions so that customers are not charged for items that they did not purchase. This type of problems can be solved through data science by applying machine learning techniques. It deals with modelling of the dataset using machine learning with Credit Card Fraud Detection. In machine learning the main key is the data so modelling the past credit card transactions with the data of the ones that turned out to be fraud. The built model is then used to recognize whether a new transaction is fraudulent or not. The objective is to classify whether the fraud had happened or not. The first step involves analyzing and pre-processing data and then applying machine learning algorithm on the credit card dataset and find the parameters of the algorithm and calculate their performance metrics.

Abstract: Drastic change in climatic conditions is a very big and challenging task for people around the globe. Most of the biological, constructional, transportation and agricultural sectors get affected due to uneven weather conditions, i.e. flood, rainfall, drought, etc. As part of the weather system, rainfall being most prominent phenomena, its rate is treated as one of the most important variables. Meteorological scientists try to identify the parameters of the atmosphere such as temperature, sunshine, cloudiness and humidity of the earth by applying conventional techniques and developing a prediction model. These days, Machine Learning (ML) techniques are more evolving and give more accurate results than the traditional approaches. ML is a subset of artificial intelligence (AI) which is used in this paper for predicting the next day's rainfall from the past 10 year’s weather dataset . This paper presents the ML classifiers such as Logistic Regression (LR), Decision Tree (DT), Random Forest (RF),and Extreme Gradient Boost (XGB) to predict the rainfall of the next day. This is followed by sequential stages of data visualization, training, testing, modelling, and cross-validation. The evaluation metrics like Area under the Receiver Operating Characteristic (AUROC) curve, recall, accuracy, precision, and Cohen kappa are used to check the performance of ML algorithms.

Abstract: Detecting COVID-19 is a difficult task for medical professionals these days due to its rapid spread. To overcome this problem, various techniques and detection methods to control the spread of COVID-19 are used. CT (Computed Tomography) Scan and X rays are currently used in the detection of COVID-19. This type of diagnostic method is accurate and fast and can be used along with normal covid-19 testing methods. The normal covid-19 testing methods such as RT-PCR method requires a radiologist to detect the disease. Therefore, it is important to implement a system to detect the corona virus automatically as an alternate quickly. This is intended to help doctors detect computed tomography (CT) images and X-Ray images of patient’s lungs infected with COVID-19. In this proposed system a Deep Learning algorithm which uses four Convolutional Neural Network (CNN) models: InceptionV3, ResNet50, Xception and VGG16 are used. These convolutional neural network models are pre-trained and we used the dataset obtained from open source which contains CT scans and X-Rays to retrain the model for the detection of Covid-19. The combined models are used for the prediction of images given by the user in a web-based prediction method. Thus, the suggested hybrid algorithm is effective for predicting images as covid or non-covid.

Keywords: Hybrid Model, CT scans and X-rays, Deep Learning Model, and CNN.

Abstract: Identification and machine-readable travel documents, or IDs and MRTDs, are used to identify and authenticate identities in a variety of contexts, including the crossing of borders, civil applications, sales and purchase portals, and access to transaction processing systems. Criminal attacks on ID verification systems now concentrate on falsified copies of real papers and the alteration of facial images because these security mechanisms are tough to go around. Governments and producers of IDs and MRTDs must constantly develop and enhance security measures if they want to lessen hazards associated with this fraud problem. In light of this, we provide the first effective steganography technique, StegoFace, which is tailored for the printing of facial photos in standard IDs and MRTDs. A Deep Convolutional Auto encoder can hide a secret message in a face portrait, creating the stego facial image. A Deep Convolutional Auto encoder can also read a message from the stego facial image, even if it has been printed and then taken with a digital camera. Together, these two components make up the end-to-end facial image steganography model known as StegoFace. In terms of perceptual quality, facial images encoded using our StegoFace method perform better than images created using the stega stamp. Results from the test set's peak signal-to-noise ratio, concealing capacity, and imperceptibility are used to gauge performance.

Keywords: StegoFace, Fake user, Deep convolutional Auto encoder, StegaStamp

Abstract: Life relies on competent functioning of heart, because heart is critical a part of our body. If function of heart isn't suitable, it'll affect the opposite body parts of human like brain, kidney etc. Heart condition could be a disease that effects on the function of heart. There are number of things which increases risk of cardiovascular disease. At this days, within the world cardiopathy is that the main reason behind deaths. the planet Health Organization (WHO) has expected that 12 million deaths occur worldwide, each year because of the guts diseases. Prediction by using data processing techniques gives us accurate results of disease. IHDPS (intelligent cardiopathy prediction system) can discover and extract hidden knowledge related with heart condition from a historical cardiopathy database. some types of heart condition are cardiovascular diseases, heart failure, coronary cardiovascular disease and Stroke. Stroke may be a sort of heart disease; it's caused by narrowing, blocking, or hardening of the blood vessels that attend the brain or by high force per unit area.

Abstract: In current era of technology, Health record monitoring and treating the patients on-time has gained wide range of scope. Utilizing the blockchain technology, is has gained significant attention to enable remote patient monitoring. Rapid development of Healthcare 4.0 using Cloud Computing to access medical records and operations remotely has gained deliberation from the technical community from a smart city perspective. As Healthcare 4.0 enhances the healthcare experience, it successfully improves the quality, flexibility, productivity, cost-effectiveness, and dependability of healthcare services. The Internet of Health Things, medical Cyber-Physical Systems, health cloud, health fog, big data analytics, machine learning, blockchain, and smart algorithms are all integrated and used. Storing, accessing, managing these healthcare sensitive data is very important. Hence there is a need for a platform that enables to achieve secure medical data storage, sharing, and accessing. Our work mainly concentrates on the modules that are utilized to provide the treatment and maintain health record of a patient in a timely and reliable manner by the healthcare professionals using blockchain technology.

Keywords: Healthcare, blockchain, managing, flexibility, health record

Abstract: Connected devices, sensors, and mobile apps make the retail sector relevant tested for big data tools and applications.We investigate how big data is, and can be used in retail operations. Based on our state-of-the-art literature review, we identify four themes for big data applications in retail logistics: availability, assortment, pricing, and layout planning. Our semi-structured interviews with retailers and academics suggest that historical sales data and loyalty schemes can be used to obtain customer insights for operational planning, but granular sales data can also benefit availability and assortment decisions can be used for demand forecasting and pricing. However, the path to exploiting big data is not a bed of roses. Challenges include shortages of people with the right set of skills, the lack of support from suppliers, issues in IT integration, managerial concerns including information sharing and process ntegration, and physical capability of the supply chain to respond to real-time changes captured by big data. We propose a data maturity profile for retail businesses and highlight future research directions. Association Rules is one of the data mining techniques which is used for identifying the relation between one item to another. Creating the rule to generate the new knowledge is a must to determine the frequency of the appearance of the data on the item set so that it is easier to recognize the value of the percentage from each of the datum by using certain algorithms, for example apriori. This research discussed the comparison between market basket analysis by using apriori algorithm and market basket analysis without using algorithm in creating rule to generate the new knowledge. The indicator of comparison included concept, the process of creating the rule, and the achieved rule. The comparison revealed that both methods have the same concept, the different process of creating the rule, but the rule itself remains the same.

Keywords: Big data; retail operations; maturity;availability; assortment; replenishment; pricing; layout; logistics.

Abstract: The agricultural production mainly aims to generate high yield for the crops.Prediction of the crop on a global scale and regional scale is highly important for the agriculture management sector, crop farmers, food trade policies and carbon cycle research. To maintain the high demand and secured level of food chain supply to the people, the prediction of crop yield is a national priority for the government. To get most crop yield at minimum value is one of the primary goals in agriculture. Detecting and dealing with troubles related with crop yield indicators in early stages of the rural field can give benefits in expanded yield and elevated earnings too. By reading weather styles of a specific location, massive-scale meteorological phenomena will have a completely green impact on agricultural production. The crop yield predictions can be utilized by farmers to reduce losses when negative conditions may occur. Also, predictions may be used to maximize crop prediction while there is favourable situation for farming. We are currently developing an automated yield estimation system that optically estimates crop yield in orchards during various stages of growth. Instead of using traditional machine vision, we build on recent advances in support vector machine (SVM) to provide results about crop details with improved accuracy rate.

Abstract: With the exponential increase of social media druggies, cyberbullying has been surfaced as a form of bullying through electronic dispatches. Social networks provides a rich terrain for bullies to uses these networks as vulnerable to attacks against victims. Given the consequences of cyberbullying on victims, it's necessary to find suitable conduct to descry and help it. Machine literacy can be helpful to descry language patterns of the bullies and hence can induce a model to automatically descry cyberbullying conduct. This paper proposes a supervised machine literacy approach for detecting and precluding cyberbullying. Several classifiers are used to train and fete bullying conduct. The evaluation of the proposed approach on cyberbullying dataset shows that Neural Network performs better and achieves delicacy of92.8 and SVM achieves90.3. Also, NN outperforms other classifiers of analogous work on the same dataset. This chapter introduces cyberbullying and does so in a way to help the anthology question “ delineations “ and understand the difficulties in this area of exploration. There's no widely agreed description of cyberbullying this chapter explores the multiple styles of cyberbullying, exercising exemplifications from the author's interviews with youthful people and published cerebral exploration.

Keywords: Cyberbullying, Fake stoner, Machine literacy, Networking

Abstract: In today’s world most of the people use internet, technology, social networking and result in huge amount of tourist data like hotels, restaurants, transport, heritage, tourist event, etc. The main reason for production of large amount of tourist data is Online Travelling Agency. However, Web Search engine provide list of possibilities to tourist but it is very difficult to find best one. Web search engine slows down the selection of best place and create noise. The result of web search engine confused the tourist. There are various recommendation systems are developed to suggest or assist tourist to plane the trip and help to find the information they are looking for. We present a recommendation system which is the combination of various recommendation system used in the field of tourism. The main objective of this work is then to contribute to the design of tourism recommender systems by proposing a framework that clarifies how the hybrid recommendation process works. The proposed system goes beyond a list of recommended tourist attractions and can be seen as a planner that aims to build a complex and detailed program of a multiday visit. Key words: recommender system, Content based filtering, Collaborative filtering, Demographic filtering, Social filtering.

Abstract: Evolution of technology has always been endeavored with making daily life simple. One of them is the visually impaired who have to rely on others for travelling and other activities. This paper aims at providing one such theoretical model which incorporates the latest technologies to provide efficient and smart electronic aid in the jacket and stick to the blind. We have used ultrasonic range finder circuit for detection. Panic situations will be sent as an SMS alert to registered mobile numbers. The basic objective of the system is to provide a convenient and easy navigation aid for unsighted which helps in artificial vision by providing information about the environmental scenario of static and dynamic objects around them. According to World Health Organization (WHO) study, 90% of the info to the human brain is sent through eyes alone. In this paper, we proposed an efficient, reliable and low-cost wearable jacket for the people suffering from visual impaired. A smart jacket is designed by embedding the sensor on the jacket that enables the user to detect an obstacle and safely navigate. The smart jacket requires low power hence can be used for real time navigation for visually impaired people.

Keywords: Smart jacket, SMS, Navigate, Obstacle, Sensor

Abstract: Many antennas are used in modern airplanes to perform various duties, sometimes in the same frequency range. As a result, it is preferable to minimise their mutual influence as much as feasible. The Radio Altimeter (RADALT) and Spiral Antenna Module (SAM) are numerically modelled using SuperNEC to discourage" and lower their coupling levels, as in the case of a RADALT and SAM mounted on an aeroplane's tail. An intelligent tilt-rotor UAV with an integrated GPS/INS system and RADALT (radar altimeter) for automated takeoff and landing is being developed. The altitude is given through a GPS/INS integration system that is referenced to the WGS-84 ellipsoid, which is susceptible to external multipath conditions. The centimetre level accuracy of RADALT delivers the aircraft AGL (above ground level) height, which is based on ground surface parameters. Simple KF (Kalman Filter) configuration to merge altitude data from the GPS/INS integration system with the operational logic of RADALT. It is assessed using data from real-world flight tests. Key words: Software development, RADALT, Altimeter, IBM DOORS, LDRA, CMULTI.

Abstract: The aim of the project is to detect the skin tone of a person from the image. A website is created which can take images as input either from a direct webcam, system or via image address. The deep learning model integrated with this will do the processing and the results will be displayed as mild, dark or fair. A Convolutional Neural Network (CNN) algorithm was developed by using the tensorflow and whereas library to differentiate images into three respective categories and train the model. The model takes image of size is resized as input. The testing can be done either by using an image already existing in the system or directly from the search engines via the URL. The images from the internet can be fetched by using the image address link, making use of the inutile. The input images will be resized and pre-processing by whereas pre-processing and then is 2 given to the model as input for prediction. Depending on the texture the image will be classified into dark, mild or fair.

Abstract: The modern world is moving fast then our older world invention of high tech appliances has revolutionized our whole world and it has also sprouted the idea towards automation. Human lifestyles have been much easier thanks to the tv remotes we have had for our tv and many other electrical gadgets today. How many of you have considered automating home, that could allow you to use a controller to operate fan, led bulbs, as well as other electronic items in your house? Because humans have little time to manage any work, automating appliances is a quick and easy technique to manage any instrument or gadget that will perform as we want. As idea of automation is rapidly developing, we have seen a number of domestic automation technologies emerge over the years, like Home for Apple, Google Home, Amazon Echo, and zigbee automation and we can see many startups are working on it . The above technology makes our homes smarter, but they require a significant investment. So, the target of the project would be to put a efficient and pocket friendly home automation program in place.By bringing into use the day to day household items like fans and geaser this system is implemented. With the help of microcontroller and special support from IFTTT and blynk programmes the voice commands which siri recieves or alexa are translated. w hen necessary, the relays attached to appliances are controlled by microcontroller, which switches the connected device to each relay ON or OFF in accordance with the customer's request to Voice Assistants.The NodeMCU [ESP8266] microprocessor is employed, and Wireless Fidelity is used to establish connection with both the microprocessor as well as the applications. Keyword: IFTTT, NodeMCU, Relay Module, and Blynk.

Abstract: Every year the crime rate of children keeps on increasing rapidly. The safety of children is the topmost priority to their parents. Despite taking various measures regarding the children's safety, children fail to protect themselves in several endangering situations. Most likely, children who travel to school without the aid of their parents are exposed to danger. Previously, there have been several cases where children have been forgotten on the bus and eventually died due to suffocation. So in this survey, the main objective is to develop a system to acknowledge parents about the entry and exit of their children during their journey to and away from the school. To implement this monitoring system, RFID (Radio- frequency identification) and GSM (Global System for mobile communication) is used. The main unit of this system is the bus unit, which detects whether the child has entered or exited from the bus. This information will be recorded in the database maintained by the school using RFID technology. The database is further analyzed and communicated. The School unit finds which one of the children did not deboard or board the bus. Simultaneously, parents receive SMS about the entry/exit of their children from the bus

Keywords: RFID, System integration, Design in Engineering, Transportation Safety, Detection.

Abstract: It helps in dealing the issues physically challenged people in single device is a difficult job. Researches focus on dealing with one challenge but not all the challenges. The main focus is on making a versatile technique that helps physically challenged people by giving more confidence to communicate with other physically challenged people. All these functionalities are being developed and made to be available in one device.

Keywords: Tesseract OCR, pyttsx, speech to text, text to speech, Image to speech.

Abstract: Digital Image Processing is that the use of a digital computer to process digital images through an algorithm. Digital Image processing could be a Software which is employed in image processing. For example: camera work, signals, photography, camera mechanism, pixels, etc. The target of this paper is to debate History of digital image processing and its applications.

Keywords: Digital Image processing, signals, mechanism, pixels.

Abstract: As the world is moving on the FinFET technology sizes is scaling down accordingly which will benefit the logic devices as the cost for the customers will be reduced and also the area required for placing it also reduces. FinFET technologies provides significant performance boosts for both logic as well as RF/mmWave over planar technologies. FinFET technology is used in 5G technologies where we need to consider the RF performance such as mobility improvement. The path of RF technology development is to concentrate on how to detach the manufacturing of RF devices from the scaling of logic devices with the least amount of negative influence on the coexistence of RF and logic devices. In order to get technology in right aspect, we need to develop runsets to check if layers are perfectly placed without any error before manufacturing it. DRC (Design Rule Check) is an important part for checking the BEOL (Backend-Of-The-Line) configuration. By maximising design productivity and acting as a doorway to the foundry where the integrated circuit (IC) will be manufactured, a well-made physical design kit (PDK) can help an integrated circuit (IC) designer achieve that goal.The main outcome of this work is to develop a runset which will provide an error free circuit design which will be beneficial for RF and the logic devices.

Keywords: DRC, Via, 5G Technology. FinFET.

Abstract: Attackers can now fabricate pictures and movies more easily because to advances in media production tools. A video downloaded from a social media can be used to make forgeries in real time.Although state-of-the-art Detecting fake photos and videos is a difficult task. This article use a capsule network to detect several types of spoofing attacks, ranging from replay attack employing graphics or scanned movies to deep convolutional neural network-generated bogus videos. It broadens the scope of capsule networks' use beyond their initial intent. We are working on the creation of a model that will be able to solve complex graphics problems. It will involve image datasets for training and testing. Terms used – capsule network, computer generated video, forgery detection replay attack

Abstract: Cancer has become a nightmare in the past few years with an increased mortality rate amongst those diagnosed with it. Early cancer detection was not possible in the past years due to a lack of knowledge, research, and slow medical development. With the changing world and growing economy, now it is possible and has become a boon to the living. Here we use artificial intelligence. An area where the system replicates human actions understands and performs computations based on its training. Deep learning is another branch of AI. Neurons in humans transmit information to the brain using electrical impulses and the brain computes the information. Similarly, in AI, we have artificial neurons that analyze the trained data and store this information. It studies this trained data and remains quiet till a new attribute is given as the input and has to be classified. Based on the previous knowledge, it classifies the new attribute and performs the operation. This is called by the name instance-based learning. AI is a field that has emerged and is growing rapidly across the world.

Abstract: Object detection followed by tracking structure which incorporates deep learning concept recognized as the Faster Regional Convolution Neural Network(Faster R-CNN) which is introduced and applied in order to detect accidents automatically when CCTV is being watched which is done in following three ways (1) Wrong-Direction (2) Stop (3) Person out of vehicle. This mechanism accepts a video input file as input which results the identification of different objects and each object identified with unique number. Thus this video is split into number of images and each image is recognized as a frame .These frames are then compared with previous frames using R-CNN algorithm to find out the accident image. This accident detection mechanism can find out accidents with minimal time. The proposed algorithm improves the results based on the quality of the video input.

Keywords: Faster R-CNN; CCTV; Object Tracking; Object detection; Deep learning; Image; Video

Abstract: Present Web Security is a serious concern. We learn with great regularity that websites, databases, and online services become inaccessible due to denial of service attacks or display modified information on the user or organization. In other high-profile incidents, millions of passwords, email addresses, and credit card information were leaked into the public, exposing users to both personal humiliation and financial risk. Web security is designed to avoid these kinds of attacks. Web Security is characterized more formally by the act of ensuring websites from unauthorized access, usage, alteration, destruction, or disruption. The higher the protection, the more trust users put on the application. Security plays a key role in achieving success for the application, especially in Web services. This essay will therefore discuss how users use cryptography today, the encryption concepts for databases, and the encryption concepts for the security of online service, thus understanding the issue of web system security.

Keywords: Protection,Cryptography,WebServices,Database

Abstract: Tree physiology and condition are closely correlated with the immediate environment and therefore is linked to climate effects in that environment. Automatic seed, plant disease identification and recognition tools have proved to be a valuable source of data that assist decision making in farms. Artificial intelligence tools like Deep learning and Convolutional Neural Network (CNN) are gaining popularity in this field as they provide optimum solution for plant disease identification. Earlier, pest detection was done by manual observation. This method is arduous and prone to error. Several plant diseases cannot be recognized by bare human eyes. Because early disease occurrences are minute in nature. At the same time due to fear of attack of pest/disease, farmer uniformly sprays pesticides/fertilizers in whole farm which may lead to damage of soil as well as plants and also infected to humans as well. In order to improve the quality of production and yield in plants, it is essential to identify the symptoms in their initial stages and treat the diseases. The crop stress index is calculated to indicate plant water status using ambient temperature. In the end we are going to implement this process to prevent the human lives from harmful effects caused by pesticides.

Abstract: Cryptography secures the data during the interaction between different systems. The attackers may use the opportunities to attack the data within the database. Therefore, the security of image is of high importance. In this idea, a private image is bifurcated into two shares of images and these images to be displayed when the two share images are available together; photos of sole share cannot reveal the identity of the actual image. To achieve this, Visual Cryptography is used. There are various dimensions on which VCS performance relay, i.e., accuracy, brightness, pixel widening, security, computer complexity, productive sharing is logical or pointless, a kind of private image. This process encrypts a private image into stocks so that it can collect a sufficient number of shares produces a private image. This project uses VC of coloured images in a biometric application.

Keywords: Biometrics; Visual Cryptography; VCS; Private Face Image.

Abstract: As a side effect of increasingly popular social media, cyberbullying has emerged as a serious problem afflicting children, adolescents and young adults. Since the textual contents on online social media area highly unstructured, informal, and Often misspelled, existing research on message level offensive language detection cannot accurately detect offensive contents. Here we design a framework called Lexical Syntactic Feature (LSF) architecture to detect offensive contents and identify potential offensive users in social media. We distinguished the contribution of profanities and obscenities in determining offensive content and introduce hand authoring syntactic rule in identifying name calling harassments. In particular we incorporated a user’s writing style, structure and specific cyberbullying contents as features to predict the users capability to send out offensive content. Results from the experiments shows that the LSF framework performed significantly better than existing methods in offensive content detection.

Keywords: Cyberbullying, Offensive ,Lexical syntactic feature, detection.

Abstract: When compared to people living in rural areas and small towns, medical facilities in major cities and towns are far more easily accessible. The process of getting medications from medical stores takes time, and these outlets might not always be open. Consequently, this might be seen as a manual process. One method that helps cut down on time wasted is an automatic pill dispenser. People often find it difficult to go to hospitals because of inadequate transportation options. Seniors and others with physical limitations have a tough time travelling and get tired from waiting in line for a doctor's appointment. Keywords- Medicines, RFID Reader, Raspberry pi, pi camera.

Abstract: Fake News is one of the major problems in the current situation. Fake News has capacity to change pinions, facts and can act as the strongest weapon in today's society. Our project uses NLP techniques for detecting the 'fake news', that is, misleading news stories which come from the non-reputable sources. The data science community has responded by taking actions against the problem. It is impossible to determine a news as real or fake accurately. So, the proposed project uses the datasets that are trained using Count Vectorizer method for the detection of fake news and its accuracy will be tested using machine learning algorithms.

Keywords: Misleading news, NLP techniques, Count Vectorizer, Train-Test split, Machine learning algorithms

Keywords: Sonic Interaction, VR, Spatial Audio, Auralization

Abstract: Depression is a mood disorder that causes a persistent feeling of sadness and loss of interest . so called major depressive disorder or clinical depression, it affects how you feel think and behave and can lead to a variety of emotional and physical problems. You may have trouble doing normal day-to-day activities, and sometimes you may feel as if life isn’t worth living. Early detection and treatment of depression are essential in promoting remission, preventing relapse, and reducing the emotional burden of the disease. Current diagnoses are primarily subjective, inconsistent across professionals, and expensive for the individual who may be in dire need of help. Additionally, early signs of depression are difficult to detect and quantify. These early signs have a promising potential to be quantified by machine learning algorithms that could be implemented in a wearable artificial intelligence (AI) or home device. This effort addresses an automated device for detecting depression from acoustic features in speech. The tool is aimed at lowering the barrier of entry in seeking help for potential mental illness and supporting medical professionals' diagnoses. Another method of implementation is through social media data. The social media platform to be used is Twitter. Live tweets are analysed and the model is trained. The project aims to have a dual mode of working between detection through audio and social media data.

Keywords: Machine Learning, Deep Learning, Convolutional Neural Networks, Feature Extraction, Depression Detection, Spectrogram Conversion.

Abstract: INDIA is one of the most populous countries in the World and is a fast-growing financial prudence. Also, Indian traffic is non-lane based. So the accident rate is more. It needs a traffic control solution, which are different from the other Countries. Hit and run is the common practice and these cases are increasing day by day. In this paper the solution to overcome this issue is provided. Vehicles will come with the unique RFID code and piezoelectric pressure will detect the accident and through message the GPS location is sent to the traffic in charge regarding the accident the victims and their locations. To avoid the accidents, alcohol sensor is implemented, which when detected will not start the engine.

Keywords: accident detection, GSM, GPS, NodeMCU, ATMEGA328,Traffic Management.

Abstract: This work considers the realisation of a human body implanted with biomedical sensors, operating wireless protocols of variable frequency, and measuring more than one physiological parameter of the body. Various nodes that are linked together to form a network of biomedical or other sensors placed at the nodes make up a wireless body area network. The implementation and introduction of the intra-body network were covered in our earlier publication, "Realization of Wireless Body Area Network utilising GNS3 tool for Health Monitoring," which has the DOI 10.17148/IJARCCE.2018.7459. The redistribution of paper routes and BGP will be discussed in this article, and efforts will be made to mimic them using the GNS3 tool. The use of a routing protocol to advertise routes that are learned by some other means, such as by another routing protocol, static routes or directly connected routers can be referred to as route redistribution. The moral behind route redistribution [1] is the content of this paper and its implementation in the constitution of the Body Area Network. Enhanced Interior Gateway Protocol which is an Interior Gateway Protocol is the protocol of choice for our project. The use of EIGRP in simulation of the inter-body network is being propounded by us. Various Routing policies will be simulated in the paper.

Keywords: Enhanced Interior Gateway Routing Protocol (EIGRP), realize a typical ISP network using a Network Simulation Tool, Redistribution, internal BGP, external BGP, Routing policies, variable frequency and 3-way handshaking.

Abstract: Side friction factors are defined as all those actions related to the activities taking place by the sides of the road and sometimes within the road, which interfere with the traffic flow on the travelled way. They include but not limited to pedestrians, bicycles, non-motorised vehicles, parked and stopping vehicles. These factors are normally very frequent in densely populated areas in developing countries, while they are random and sparse in developed countries making it of less interest for research and consequently there is comparatively little literature about them. The objective of this thesis is to analyze the effect of these factors on traffic performance measures on urban roads. To carry out this work, a research design was formulated including specific methods and prescribed limitations. An empirical case study methodology was adopted where Dar-es-salaam city in Tanzania was chosen as a representative case. The scope was limited to include only road-link facilities. A sample of these facilities including two-lane two-way and four-lane two-way roads were selected and studied. The study was conducted in two parts, of which each involved a distinctive approach. Part one involved a macroscopic approach where traffic and friction data were collected and analyzed at an aggregated level, whereas part two involved a microscopic approach where data of individual frictional elements were collected and analysed individually. Data collection was mainly performed by application of video method, which proved to be effective for simultaneous collection of traffic and side friction data. Data reduction was conducted chiefly by computer, using standard spreadsheet and statistical software packages, mainly SPSS and some computer macros.

Keywords: Motorised, friction, road.

Abstract: Passenger car units (PCU) are used to represent the effects of varying mixed vehicle types on traffic stream. In this paper the required data is collected at eight sections of main highways of Nepal using a digital video recorder which eventually analyzed the traffic characteristics and PCU values was calculated. The study found traffic composition of Bus, Truck, LCV and Car are increasing with the increase in carriageway width but the composition of volume is found to be highest in smaller carriageway width. The speeds of all categorized vehicles are increasing linearly with the increase in carriageway width. It is found that PCU values obtained for motor cycle from all sections are smaller than the values given in NRS and for Bus, Truck, LCV found higher than the value given in NRS 2070 .This study has shown the impact of lane width on the PCU for different categories of vehicles on a Highways. It is found that the PCU for a vehicle type increases with increasing carriageway width.

Keywords: Traffic composition, Speed, Passenger Car Units

Keywords: Brain Computer Interface, Brain sense, Arduino, Home automation, Home appliances

Abstract: Skyline queries are one of the most commonly used query operators for locating query results that only return data items whose dimension vector is not dominated by any other data item in the database. Skyline queries have been included into various sorts of databases, including complete, incomplete, and uncertain, due to their utility and ubiquity. In the early phases of a knowledge-discovery process, the skyline query and its variant queries are useful functions. The skyline query and its variants identify a collection of important items that outperform the dataset's other common objects. Such knowledge-discovery queries must be computed in parallel distributed systems in order to manage huge data. We will use typical datasets in this research to parallelize skyline computations utilising various strategies such as parallel data pre-processing, parallel skyline computation using multithreading, and multiprocessing. The ultimate goal is to reduce response time using parallel computation. We shall be able to give the approach for efficient, parallel skyline calculation near the end of the paper.

Keywords: Skyline query, Preference queries, Skylines, SQL, Algorithms, Database, Multithreading, multiprocessing.

Abstract: Skyline queries are one of the most commonly used query operators for locating query results that only return data items whose dimension vector is not dominated by any other data item in the database. Skyline queries have been included into various sorts of databases, including complete, incomplete, and uncertain, due to their utility and ubiquity. In the early phases of a knowledge-discovery process, the skyline query and its variant queries are useful functions. The skyline query and its variants identify a collection of important items that outperform the dataset's other common objects. Such knowledge-discovery queries must be computed in parallel distributed systems in order to manage huge data. We will use typical datasets in this research to parallelize skyline computations utilising various strategies such as parallel data pre-processing, parallel skyline computation using multithreading, and multiprocessing. The ultimate goal is to reduce response time using parallel computation. We shall be able to give the approach for efficient, parallel skyline calculation near the end of the paper.

Keywords: Skyline query, Preference queries, Skylines, SQL, Algorithms, Database, Multithreading, multiprocessing.

Abstract: Depression is the most common mood disease in the world, with serious consequences for one's well-being and functionality, as well as substantial personal, familial, and societal consequences. Early and effective diagnosis of depression symptoms could provide numerous advantages for both physicians and those who are affected. The goal of this study was to create and evaluate a methodology that could detect visual symptoms of depression and help clinicians make judgments. The field of automatic depression evaluation based on visual clues is fast expanding. The current comprehensive assessment of existing methodologies focuses on image processing and machine learning algorithms, as documented in over sixty articles over the last ten years. The visual signs of depression, various data collection methodologies, and available datasets are all summarized. The review discusses visual feature extraction methods and algorithms, dimensionality reduction, classification and regression decision methods, and various fusion methodologies. A quantitative meta-analysis of published data is given, based on performance criteria that are robust to chance, to indicate general trends and important unresolved concerns for future investigations of automatic depression evaluation using visual cues alone or in combination with visual cues. The proposed work also used deep learning to forecast the level of depression based on current input of face photos.

Keywords: Convolutional Neural Network, Deep Learning, Dataset, Depression.

Abstract: This study suggests an electrocardiogram (ECG) analysis and heart illness detection artificial intelligence of things (AIoT) system. A cloud database, a user interface on a smart device application (APP), front-end IoT-based hardware, and an AI platform for heart illness diagnosis are all included in the system. The wearable ECG patch with an analogue front-end circuit and a Bluetooth module, which is the front-end IoT-based hardware, can detect ECG signals. The APP on smart devices may identify diseases in real time and classify odd signals in addition to displaying users' real-time ECG readings. The cloud database will receive these ECG signals. Each user's ECG readings are stored in the cloud database, creating a big-data set that an AI programme may use to identify heart problems. The convolutional neural network-based approach that this study suggests has an average accuracy of 94.96 percent. The Ministry of Health and Welfare's Tainan Hospital provided the ECG dataset used in this investigation. Additionally, a cardiologist additionally carried out signal verification.

Keywords: Arrhythmia, atrial fibrillation, convolutional neural network, electrocardiogram, artificial intelligence of things, wearable device, application, cloud server.

Abstract: In current electronic world, further losses are impacted by cybercrime. Digital world are completely open and it's incredibly easy to concentrate to get to data or cash related information from any existent, open and private associations and so on, since, endlessly web are open, taking data on the unstable medium of channel is outstandingly straightforward. Consequently, guaranteeing bad geste data needed advance layers of safety counter to the cybercrime. Data evidence is one of the first noteworthy systems to staying down from data from digital shamefaced parties. In this data evidence procedure, cryptology has introductory influence against the cybercriminal on the unstable correspondence channel. It gives data security, sort out security, perceptivity and character association to get to the data approved work force. Colourful open and private crucial systems are proposed for getting data, anyhow there are as yet a piece of challenge live in this substance. Utmost unmistakable encryption methods, for illustration, RSA, Elliptic wind, DES and AES are working. In this undertaking about new cryptography system averting the cybercrime in view of the blockchain.

Keywords: Alert system, Crime Data, Safe area, Security, AES algorithm, Blockchain, Etc.

Abstract: Phishing websites that anticipate to take the victim’s confidential data by diverting them to surf a fake website page that resembles a sincere to goodness one is some another type of criminal activity through the internet and its one of the especially concerns in numerous areas including e-managing an account and retailing. Detecting phishing sites is a complex and unpredictable process involving numerous variables and criteria that are not stable. Using Extreme Learning Machines, we proposed an intelligent model for detecting phishing web pages. There are different types of web pages with different features. Therefore, we must use a specific set of features on web pages to protect against phishing. A machine learning model was proposed to detect phishing web pages. To detect phishing web pages, we proposed a machine learning model. This study aims to detect phishing URLs and narrow down the best machine learning method based on accuracy, false-positive rate, and false-negative rate. Phishing, Feature Classification, Random Forest Classifier, and other terms are used in this study.

Keywords: Phishing attack, Machine Learning, Random Forest, Feature Classification, URL.

Abstract: The heart is an important organ in living things. Heart-related disease diagnosis and prognosis calls for greater research precision, excellence, and correctness because even the smallest error can cause fatigue issues or individual death, there are many heart-related deaths are becoming more common, and their number is growing day, exponentially. A illness awareness prediction system is absolutely necessary to address the issue.Machine learning, a subset of artificial intelligence (AI), offers excellent assistance in making predictions about any form of event using data from real-world occurrences. In this study, utilising the UCI repository dataset for training and testing, we measure the accuracy of machine learning methods for predicting cardiac disease. These algorithms include k-nearest neighbour, decision tree, linear regression, and support vector machine (SVM). The greatest tool for implementing Python programming is the Anaconda (Jupytor) notebook, which has a variety of header files and libraries that improve the accuracy and precision of the task.

Keywords: supervised; unsupervised; reinforced; linear regression; decision tree; python programming; jupytor Notebook; confusion matrix;

Abstract: Telecom industry has gained a huge growth in the last two decades. Because of the availability of a lot of options, many telecom companies are facing the problem of customer churn in the recent years. Because of the advancement of and indispensable need for internet, customers can easily change from one company to another. It may affect the profits of the Telecom companies if they don’t pay enough attention for customer churn. To pay attention to the customer churn, the Telecom companies should be able to predict which customers are likely to leave the company. Manually predicting this is almost impossible. With the help of machine learning algorithms, we can try to predict the customers who could possibly switch over. Once the companies know if a customer is going to churn, they can try to retain those customers through various strategies.

Keywords: Random Forest, Decision Tree, SMOTEENN, ML, Churn

Abstract: The use of online tools for a faster generation of essays or sentences has always been very important. We propose a one-destination website that provides users with the facilities to be able to choose between Essay generation, next-word prediction, or paraphrasing. The tools used typically help users not just with typing but also with grammatical errors, better sentence construction, and quicker more efficient outputs. Our project uses GPT, web scraping, and transformers. Pre-trained transformers have attained a state-of-the-art performance for various NLP tasks. We mainly focus on providing users with a one-stop destination for their needs related to the topic. The website saves users time having to find results for their needs. Moreover, the site helps reduce plagiarism while still providing a good result for the user’s needs.

Keywords: Latent Semantic Analysis, Paraphrasing, Next word prediction, Natural language processing, BERT, transformers, LSTM, Character prediction.

Abstract: In this paper, we introduce a new fine-grained two-factor authentication access control system for web-based cloud computing services. Specifically, in our proposed 2FA access control system, an attribute- based access control mechanism is implemented with the necessity of both a user secret key and a lightweight security device. As a user cannot access the system if they do not hold both, the mechanism can enhance the security of the system, especially in those scenarios where many users share the same computer for web-based cloud services. In addition, attribute-based control in the system also enables the cloud server to restrict the access to those users with the same set of attributes while preserving user privacy, i.e., the cloud server only knows that the user fulfills the required predicate, but has no idea on the exact identity of the user.

Keywords: web-based cloud services, two-factor authentication, 2FA access control system, cloud storage

Abstract: Due to the advancement of wireless communication, there are several online security risks. The importance of intrusion detection systems (IDS) in computer and network security cannot be overstated. The experiment dataset in this research was the KDDCUP'99 (Knowledge Discovery Dataset) intrusion detection dataset. Due to intrusion detection's fundamental properties, there is still a significant imbalance between the classifications in the dataset, which makes it more difficult to apply machine learning to intrusion detection efficiently. IDS techniques come in a wide variety and yield results with varying degrees of precision. This calls for the creation of an efficient and reliable intrusion detection system. In this paper, a method for creating effective IDS that makes use of the random forest classification algorithm and principal component analysis (PCA) is proposed. While Random Forest (RF), an ensemble classifier that outperforms other standard classifiers for the accurate classification of attacks, PCA will assist in organizing the information by reducing its dimensionality. Together with the Confusion Matrix, a performance evaluation tool, we have also employed other approaches for model evaluation and selection, including as accuracy, precision, recall, and f-score.

Keywords: IDS, Knowledge Discovery Dataset, PCA, Random Forest

Abstract: A Face Recognition System is an application of computer vision and Image processing which is capable of performing two major tasks of identifying and verifying a person. The system proposes a solution for implementing a face recognition-based attendance system using Python, OpenCV and KNN. A GUI for the project is also designed using PyQt, which is a Platform to create GUI based Programs.It is designed to capture the images from live camera feed within a given time frame and recognize known faces from the images and mark their presence.It uses CSV file to store the attendance report for the session which can then laterbe exported. By automating the attendance process the system thereby reduces the time taken and manual work involved in traditional attendance methods.

Abstract: Programming today demands to be typed manually as well as with the option to write code literarily. As a result, both of these have placed people with disabilities in an unquestionably difficult situation when it comes to understanding and using coding. Few of the technologies that enable computer use by people with disabilities are the specific solutions for helping people with disabilities with various coding issues. The goal of this project is to build a portable Web interface, especially for people with disabilities to learn how to code. A web interface is planned, with a focus on speech recognition, semantic analysis, and people working together with coding tools to help physically and visibly disabled people. It is a strategy for implementing a superior method of aiding and persuading handicapped people, including the most recent developments in information innovation and computer programming viewpoints. Here, rather than adopting the more current syntax that software professionals use, we are concentrating on the C language because it is seen as being fundamental. This project helps all people with physical disabilities who are interested in coding by demonstrating programming concepts and motivating people to use programming models to develop their learning capacity and coding hypothesis. The individual is led step-by-step by their abilities, using oral orders to act immediately at any time.

Keywords: Vocal Programming , Programming tool , Speech Recognition ,Multimodal Interfaces , Voice I/O , Voice Accessibility

Abstract: This paper presents a simulation based approach to analyse EMI/EMC levels or effect on a vision sensor PCB enclosed in a metallic enclosure. Specifically EMI analysis is done based on radiated emission test, Conducted emission test and ESD test. In this project focus is given to Radiated emission test based on CISPR 25 standard, which is standard to carry out EMI/EMC tests. Apart from electromagnetic tests, thermal analysis of the board is also performed using simulation methods. For simulation purpose ANSYS tools are used here. Proper analysis of Radiated emission and thermal aspects would ensure reliability of the board. EMI effects can lead to failure and misfunctioning of the components, whereas thermal effects can lead to failure of interconnects and increasing overall temperature of the environment.

Keywords: Electromagnetic interference(EMI), ANSYS, CISPR 25, Thermal, Radiated emission, Conducted emission, ESD.

Abstract: The increase in online services like medical consultancy on internet, business trades taking place in the web so on and so forth, in many means have proliferated the need for abusive content detection and its prevention because of the vast data flowing in the web. In addition to that, the multiplied cyber crimes have been concerning the communities that are looking to enlarge their business by making it accessible to a wide range of communities by moving to the global networks. Abusive content is against humanity and might lead to mental disabilities like depression and many more severe issues amongst the victims, so it is a basic need to prevent the abuse and make sure the data is secure that is provided by clients belonging to various sections of the society. This project involves detecting multiple cases of abuse in the comments, tweets, and messages. It detects positive and negative sentences.

Abstract: This paper presents solar powered motor vehicle with the efficient charging system. This solar vehicle is used as one of the cardinal energy-saving vehicles where the application of renewable energy meets sustainable energy demand with reduction of fuel cost and purification of the atmosphere. The energy for the vehicle will be supplied by Solar panel (10W). For sufficient energy management, the charging system is used. The charging is independent of vehicle movements. When Solar panel will get sunlight, energy will be supplied to Charge controller. The battery which is connected to the charge controller will get charged via the energy received from the panel. The motor driver IC and the Bluetooth module will get power from the Arduino which gets power directly from the charge controller. Hence, the vehicle will move. Index Terms: Solar panel, Light-Weight Vehicle, Renewable Energy, Lithium-Ion battery, Arduino, Solar Charge Controller.

Abstract: Facial expression recognition is an essential issue in the field of human and computer interaction. The objective of this paper is to deal with the problem of failure for the feature points on face based on actives appearance model (AAM) in fitting with different users, A high quality AAM alignment results rely on appropriate selections of initial positions, In our method, we apply partial AAM fitting on mouth and eyes, To obtain stable partial AAM, we use multi-level optical flow to determine initial position of facial feature models. Then we use the Eigen-faces dynamic face recognition system to recognize different users and select the trained fitting model in recognition with Eigen-face dynamic face recognition. In this paper, we have demonstrated our approach which can accurately recognize different users facial expressions and emotional variation.

Abstract: With the rising usage of room related information, the interest for spatial enormous information sharing and exchanging is developing quickly, which advances the rise of spatial information market. Notwithstanding, in customary information markets, the two information purchasers and information dealers need to utilize a concentrated exchanging stage which may be deceptive. Blockchain is a decentralized dispersed information capacity innovation, which utilizes the detectability and unforgeability to affirm and record every exchange, can tackle the inconveniences of the unified information market, in any case, it likewise presents the issues of safety and security. To address this issue,in this paper,we propose a blockchain-based spatial information exchanging structure with Trusted Execution Environment to give a believed decentralized stage, including information stockpiling, information question, information estimating and security processing. In view of this structure, a spatial information exchanging show framework was carried out and its plausibility and security were checked.

Keywords: blockchain,trading,ecosystem

Abstract: After the breakout of the worldwide pandemic COVID-19, there arises a severe need of protection mechanisms, face mask being the primary one. The basic aim of the project is to detect the presence of a face mask on human faces on live streaming video as well as on images. We have used deep learning to develop our face detector model. The architecture used for the object detection purpose is Single Shot Detector (SSD) because of its good performance accuracy and high speed. Alongside this, we have used basic concepts of transfer learning in neural networks to finally output presence or absence of a face mask in an image or a video stream. Experimental results show that our model performs well on the test data with 100% and 99% precision and recall, respectively. Identifying a person by face is quite a trend nowadays. but here we are going to check whether the person is wearing mask or not. And then we can detect whether the person is an authorized person or an unauthorized person. We are going to use the Open-CV and CNN (Convolution Neural Network) to detect the presence of mask and to detect the person's identity. For face detection, Haar-cascade is used, for face recognition Eigen faces and fisher faces are used. When we find an unauthorized person. The system can be able to generate an alert e-mail and send it to the concerned e-mail address. And the graphs are drawn using Matplotlib library.

Keywords: covid, facemask, deeplearning, detection, email, warning

Abstract: This paper demonstrates digital security vs. private information. Digital security will be considered a world security agenda that will concern protecting states and citizens from the misuse of sensitive information. By using privacy and security measures, a corporation can manage the protection threats related to it. Moreover, it is a significant a part of the organization because it prevents financial and reputational damage for any organization. This report aimed to know the small print of digital security and also the issue associated with the identical in terms of handling private information. It sets the objectives and background to the study within the introduction section. It then went on to research the methodology section. a combination of primary and secondary research was employed in this project, using survey and thematic analysis. The research paper then collected data and analysed the identical with the assistance of MS Excel graphs and charts and peer-reviewed journals and articles. This report is ready to satisfy the digital economy and manage the digital security and privacy risk for various organizations' social and economic prosperity. This paper discusses the critical dimensional areas towards digital security to attach with the digital environment and privacy of the knowledge technologies. This paper has been articulated risk management by analysing the possible threat areas to comprehend the present situation. It came up with the knowledge that digital security has the prospect of causing significant issues while storing and securing private information.

Keywords: Digital security, encryption, RFID, risk assessment.

Abstract: Chat bots typically provide a text-based user interface, allowing the user to type commands and receive text response. Chatbots are usually a stateful services, remembering previous commands in order to provide functionality. Since the discovery of the Covid-19, it has become a global pandemic. At the same time, it has been a great challenge to hospitals or healthcare staff to manage the ow of the high number of cases. Especially in remote areas, it is becoming more difficult to consult a medical specialist when the immediate hit of the epidemic has occurred. Thus, it becomes obvious that if effectively designed and deployed chatbot can help patients living in remote areas by promoting preventive measures and reducing psychological damage caused by isolation and fear. Our idea behind this project is to present sophisticated chatbot for users, especially during unknown pandemics like Covid-19. This chatbot answers the FAQ's related to Covid-19.

Abstract: Deep learning algorithms have recently expanded their applications beyond big data analytics to include intrusion detection systems. Artificial intelligence and image processing advances are changing and challenging how people interact with digital images and video, and because of their intrinsically contentious character and the reach of contemporary society, they are intended to propagate harmful content and disinformation to millions of people. A picture may say a thousand words, but what if the photograph has been fabricated? The term "fake news" has recently gained popularity, yet with today's photo manipulation techniques, even the most vigilant eyes can be tricked. One of these areas is the use of several software such as faceapp and fakeapp to create modified media files known as deepfake. From massive data analysis to human biometric systems, deep learning algorithms are used. Due to their user-friendly characteristics, these applications are growing more popular with the general public and are employed in a range of fields including digital fraud, cybercrime, politics, and even military actions. As a result, it's critical to build detection technologies that can detect and remove this form of forgery, as well as to take a new step forward in video and audio forensics.

Abstract: So many organizations that are moving from their traditional data storage to cloud storage which provides and efficient way to access the data anywhere and anytime. We know that security in cloud computing is the emerging issue in current time. But the main problem of organization is to use cloud computing in the data security. In this paper we have proposing multi level cryptography best security system in cloud computing. We are using new hybrid approach of symmetric and asymmetric key cryptography algorithms. There are two algorithms that we are using are Data Encryption Standards (DES) and RSA Aag implementing here to provide the multilevel of encryption and description at both sender and receiver side which can increase in their security of cloud storage. As this model the security of this model gives you the transparency to the cloud user as well as cloud service provider in order to reduce the security problems. Also this model increases the data security up to a maximum extent and takes less time in processing the text files as comparing to the other existing system.

Keywords: Data Security, Hybrid cryptography, cryptography, Cloud computing, RSA, DES

Abstract: Object detection and recognition systems have gained significant interest of researchers due to vast advancement in the field of computer vision technology. Although there are number of object recognition systems implemented in past researches, there still remains a constant demand for new, better and accurate recognition systems. Current detection systems make use of classifiers to perform detection. We are implementing a machine learning model which can detect objects using the concept of Convolutional neural networks (CNNs). This enables us to detect objects with very high accuracy.

Keywords: Neural Network, Opencv, YOLO, Non-Max Suppression

Abstract: Currency forgery is a significant crime that has a negative impact on a country's finances. In banking systems, the proposed technique will be useful in detecting counterfeit money. Due to a rise in the number of counterfeit notes on the market, India is experiencing more serious issues. Various false note detecting solutions are available globally to combat this problem, however the most of them are hardware-based and expensive. This focuses on obtaining public access in order to detect counterfeit currencies. The suggested method can determine a banknote's legality by looking for certain security features including watermarks, latent pictures, security threads, and so on. Machine learning algorithms can be used to identify counterfeit banknotes. These security aspects are extracted and encoded as part of the approach. A support vector machine is used to extract security features from the input image, as well as to identify and classify them.

Keywords: Counterfeit currency, Convolutional Neural Network (CNN), Support Vector Machine (SVM), Android Application, Region of Interest (ROI), Edge Detection, Artificial Intelligence (AI), Image Processing, Machine Learning (ML), Deep Learning(DL).

Abstract: Emergency services at the time of road accidents are crucial for rescuing people in need of critical medical attention. Most road accidents lead to deaths when the victims don’t make it to the hospital in time or when their requirements aren’t met in a short period. Involving smartphones in the field of emergency medical services can boost the quality of service that the victims receive. Finding a good hospital service in an unknown geographical location is also a major challenge. A web application to deal with these issues can serve as a big leap in the field of medical service for road accidents. The proposed Web application provides the user with a list of all hospitals present in the nearest location with the help of GPS. It also gives an option to view the list of nearest blood and organ donors who are eligible. It works on finding the hospital for people who are in need of emergency in the time of an accident. This app would hence save a lot of time and effort in finding emergency medical services. This might save a lot of lives in need during the time of emergency and make the work easier for the blood donation centers and the hospitals.

Keywords: Emergency service, Eventually, emergency treatment, and medical facilities

Abstract: There is a new and better password key generation method is proposed in this paper. This algorithm is going to provide fast and secure ki generation in public Cryptography world in this case we are going to use the RSA algorithm with the help of random numbers (Dynamic keys) without waste of offline and online dictionary malware. In this new proposed algorithm is closely related to prime number generation techniques that can creating a new pair of keys between authorised users. Methods are much important for some uses, i.e. reducing password guessing and knowing probability by others in other ways. And this propose class of secure password generating method is provides safest and continuous transaction between authorised users.

Keywords: Dynamic keys, RSA, Asymmetric Key Cryptography, prime number

Abstract: CLOUD computing offers its customers an affordable and convenient pay-as-you-go service model, known also as usage-based pricing. specifically, data transfer costs (i.e., bandwidth) is also an important issue when trying to chop back costs. Consequently, cloud customers, applying a judicious use of the cloud’s resources, are motivated to use various traffic reduction techniques, specifically traffic redundancy elimination (TRE), for reducing bandwidth costs. Traffic redundancy stems from common end-users’ activities, like repeatedly accessing, downloading, uploading (i.e., backup), distributing, and modifying the identical or similar information items (documents, data, Web, and video). TRE is employed to eliminate the transmission of redundant content and, therefore, to significantly reduce the network cost. In commonest. While proprietary middle-boxes are popular point solutions within enterprises, they do not seem to be as attractive during a very very cloud environment. Cloud providers cannot experience during a technology whose goal is to chop back customer bandwidth bills, and thus don't seem to be likely to take a position in one. the increase of “on-demand” work spaces, meeting rooms, and work-from-home solutions detaches the workers from their offices. In such a working environment, The fixed-point solutions that need a client-side and a server-side middle-box pair becomes ineffective. On the choice hand, cloud-side elasticity motivates work distribution among servers and migration among data centers. Therefore, it's commonly agreed that a universal, software-based, end-to-end TRE is crucial in today’s pervasive environment. within the case where the cloud server is that the sender, these solutions require that the server continuously maintain clients’ status. We show here that cloud elasticity demand a replacement TRE solution.

Keywords: Web application, Online Admission, ERP, Online Result.

Abstract: The data in healthcare has increased in volume, intricacy and comprehensiveness. This growth leads to extensive application of artificial intelligence and machine learning in the healthcare sector. This study aims to examine the application of deep learning models and ensemble learning in diagnosis prediction. We apply Natural Language Processing techniques on medical notes to predict diagnosis. Real-life healthcare datasets, like MIMIC-2, contain tables with medical notes which can be pre-processed and used to train ML models. This paper presents an analysis of a diagnosis prediction algorithm. This facilitates the creation of autonomous medical systems which can be used to aid or act in place of healthcare professionals.

Keywords: MIMIC 2, EHR, Clinical notes, NLP, Bidirectional LSTM, BERT, Ensemble learning.

Abstract: After the boom and bust of cryptocurrencies’ prices in recent years, it has been increasingly regarded as an investment asset. Because of its highly volatile nature, there is a need for good predictions on which to base investment decisions. Although existing studies have leveraged machine learning for more accurate Cryptocurrency price prediction, few have focused on the feasibility of applying different modelling techniques to samples with different data structures and dimensional features. To predict Cryptocurrency price at different frequencies using machine learning techniques, we first download the dataset from a trusted website which keeps all the data of various cryptocurrencies then we classify various Cryptocurrencies by the dataset that is according to the available price. We extract the basic trading features acquired from a cryptocurrency exchange are used for 1 month price prediction. Machine learning algorithms including ARIMA and SVR models for Cryptocurrency’s daily price prediction with high-dimensional features achieve an accuracy of 93% and 94% respectively, outperforming more complicated machine learning algorithms. Compared with benchmark results for daily price prediction, we achieve a better performance, with the highest accuracy of the machine learning algorithm of 97%. Our Hybrid Machine learning model including Support Vector Regression and Autoregressive integrated moving average for One month’s Cryptocurrency price prediction is superior to other Machine learning methods, with accuracy reaching 97%. Our investigation of Cryptocurrency price prediction can be considered a pilot study of the importance of the sample dimension in machine learning techniques.

Keywords: Cryptocurrency, Autoregressive Integrated Moving Average (ARIMA), Support Vector Regression (SVR), Price Prediction, Bitcoin, Transactions, Accuracy, Machine Learning (ML), Deep Learning (DL).

Abstract: In the current age, the public activity of everybody has become related with online interpersonal organizations. These locales have rolled out an extraordinary improvement in the manner we seek after our public activity. Making companions and staying in touch with them and their updates has become more straightforward. In any case, with their fast development, numerous issues like phony profiles, online pantomime have likewise developed. There are no practical arrangements exist to control these issues. In this paper, I thought of a system with which the programmed ID of phony profiles is conceivable and is productive. This structure utilizes grouping strategies like Random Forest Classifier to order the profiles into phony or veritable classes. As this is a programmed location technique, it tends to be applied effectively by online informal communities that have a great many profiles whose profiles can't be inspected physically

Keywords: social media, Facebook, Random Forest Classifier, Classification, Framework, and Dataset

Abstract: This paper presents the applications and the effective use of Solar Wind Hybrid Energy systems (SWHES). The future of Energy generation depends on Solar Energy, as it the most abundant natural source of energy. Conventional power generation is going to become a difficult task in the future; it is due to the non-availability of coal. The increased per unit generation cost in the thermal power plant. The transmission power loss is also one reason. Pollutants released from the conventional power generation will affect the environment. To overcome these difficulties in future we have to depend on solar power generation. It is clean source of energy and it can transform to any source of energy with no effect on the environment. To get continuous power supply we should operate wind and solar power plants together as a single unit. By this combined mode of operation, the overall efficiency of the system increases. The combined power generation will give the continuity power supply for household applications with battery as a storage element. SWHES are more reliable to small power application. This configuration also reduces the load on the conventional power generating system with no effect on the environment.

Keywords: Hybrid Energy Systems, Solar Power Applications, Wind Power Applications, Combined Power Generation, Continuous Power Supply, SWHES.

Abstract: One of the most intriguing new developments in artificial intelligence is machine learning. There are several daily-used programs that incorporate learning algorithms. The moment to search the internet, a web search engine like Google or Bing is utilized. One of the reasons the internet functions so well is that a Google-implemented learning algorithm Microsoft has mastered the art of ranking websites. Each time Facebook is utilized, and it also identifies friends' images computer learning. Email spam filters protect users from also annoying to have to go through a ton of spam mails. In this essay, a quick summary and potential. The potential for machine learning's numerous applications has been created. The use of artificial intelligence (AI) has grown across numerous industries, including healthcare. Today, health care organizations of all sizes, types, and specializations are more interested in how artificial intelligence has developed and is assisting with patient requirements and care while also lowering costs and boosting effectiveness. This study investigates the effects of AI on healthcare administration, and problems associated with implementing AI in healthcare, as well as an analysis of numerous academic papers that employed AI models in a variety of healthcare fields, including dermatology, radiology, drug development, etc. The functions and services of healthcare have been significantly improved by the rise of machine learning (ML) and blockchain (BC). This study uses bibliometric visualization to examine how ML and BC are applied in the field of smart medicine. We list the nations with the highest output and most popular research topics.

Keywords: Artificial intelligence, Machine learning, Supervised learning, Unsupervised learning, Reinforcement learning Applications, Blockchain.

Abstract: The IoT and android-based Weather Monitoring System project is used to get Live reporting of weather conditions. This system is to provide an efficient weather monitoring system by monitoring the weather data based on Internet of Things (IoT) technology and to show the weather data by using the mobile application with quick and easy access for end users. The system uses NodeMCU microcontroller board which is used to collect weather parameters from temperature and humidity sensor (DHT22), digital Barometric Pressure Sensor (BMP280), digital Ambient Light Sensor (BH1750), Rain Sensor (FC37) and digital Air Quality Sensor (CCS811). Data collected from the sensors are then stored into the Firebase real-time database, a cloud-hosted database and mobile application are developed using Android Studio to show the real-time weather conditions in an android mobile phone.

Keywords: NodeMCU, Barometric, Ambient, Arduino IDE, Android studio, Firebase.

Abstract: Online purchasing is gradually displacing traditional shopping methods in every manner. As a result of pandemic, it has gained traction and has become the new normal. From shoes to food, everything is now available on E-cart. People like to purchase online since there are so many possibilities in each area. Users can notice the price differences between websites and, as a result, the majority people will choose the service with the lowest price. So, in order to stay ahead of the competition, business minds are constantly devising new strategies in order to maximize the profit from each sale. They come up with enticing offers in order to attract more clients. In the E-commerce industry, dynamic pricing is currently the most popular method. Consumer may find it challenging to keep up with the pricing changes that occur every 10 minutes on average. They give the illusion that the goods is on sale on that particular by doing so. We offer a basic price tracker application in this paper that tracks the price of a product and notifies the customer when the price reaches the desired level. It will get the customer pricing from the browser and act in accordance with the customer’s wish.

Keywords: MIMIC 2, EHR, Clinical notes, NLP, Bidirectional LSTM, BERT, Ensemble learning.

Abstract: Cervical cancer is the leading root of cancer demise in women, bordering on breast cancer. Cervical cancer is a strain of cancer that arises in the porta cells – beneath a scrap of the uterus that links to the quim. Profuse shears of the human papillomavirus (HPV), an erotically pass-on ulceration frolic a vital walk on the part in giving rise to the mass cervical cancer. The risk of emerging cervical cancer can be diminished by having screening tests and obtaining an antivenin that keeps safe hostile against HPV infection. Cancer determent in the focal is procured by veiling the transfiguration tract. Cervical fibrotic junctures can be espied in ternion divergent breeds, and wholly can be mutated into a tumor. As an upshot, it's decisive to canopy cervical anomalies pragmatically and have an attested undertaking to dictate if a cervix is habitual (healthy) or pigmentary. At this moment, the inspection being committed is a Pap smear test, often mentioned as a Pap test, which is a cervical screening stratagem. It scrutinizes your cervix for the showing-up of pre-cancerous or cancerous cells. The Pap scrutiny's premier snag is that like countless it cannot clinch an authentic sequel. A camouflaged Pap test proclaims deviant cells in the cervix when there are no extant. At prompt times deep learning is fetching a supplemental scathing for cancer concealing. Cervical cancer detection and classification technique deployed on Convolution Neural Network (CNN) has been tendered. Deep-learned facets are procured using the CNNs paradigm. The strategy has paraded atypical recital, witnessing the bids tack might in remitting a potent tool for cervical cancer assortment in clinical milieus. Index Term – Cervical cancer, cervix, pap smear test, convolution neural network.

Abstract: Interplanetary File System is a decentralized storage architecture that provides web3 storage API. For sharing files to IPFS we need to put files in the operating directory for generating hash which will be offered by web3.storage resulting in storage of data in the file-coin network, and its unique economic model and over 15EiB of capacity allows us free storage

Keywords: Blockchain, IPFS, file-coin, CID

Abstract: This work is originally present in “App’s Auto-Login Function Security Testing via Android OS-Level Virtualization”.Most mobile apps provide the automated login option to improve user experience despite the small keyboard's limitations. Users can avoid the hassle of having to type their ID and password again whenever an app is running in the foreground. The so-called "data-clone attack" can be launched by copying the locally stored, auto-login dependent data and installing it on the attackers' smartphones, which allows the attackers to exceed the allowed number of login devices and secretly connect into the victim's account. Verifying the consistency of device-specific properties is a natural countermeasure. The programme will block the auto-login feature and hence guard against data-clone attempts as long as the new device displays distinct device fingerprints from the old one. In this article,with VP Droid, security analysts can alter several device artefacts in a virtual phone without using user-level API hooks, including CPU model, Android ID, and phone number. The isolation method of VPDroid makes sure that user-mode apps in the virtual phone cannot identify differences between devices. We simulate data-clone attacks with 234 of the most popular Android apps using VPDroid in order to evaluate how vulnerable Android apps are to these assaults. Our tests on five distinct virtual phone settings demonstrate that all evaluated apps that do device-consistency checks, such Twitter, WeChat, and PayPal, may be tricked by VPDroid's device attribute customisation. As a zero-day vulnerability, our report has been verified by 19 vendors.

Abstract: The exhaustion of conventional resources and its effect on climate requires an urgent call for the substitute power resources to convene up the current power requirement. Solar energy is an endless, unsoiled and prospective energy source among all other nonconventional energy options. As more concentration is being done on focal point for the development of renewable energy capital globally. To ascertain their viability it is necessary to do the economic and technical assessments of these resources. This paper presents designing aspects and assessments of solar PV system based on field and actual performance. The study is based on design of solar PV system and a case study based on cost analysis of 1.0 kW off-grid photovoltaic energy system installed at Jamia Millia Islamia, New Delhi (28.5616°N, 77.2802° E, and about 293 m above sea level) India. Both monthly and weekly costs of energy produced by the 1kW PV system have been calculated. In addition, the solar PV 1kW system can give internal rate of return of about 1.714% on investment. Based on assumptions used in this study, solar 1kW PV system of Rs 0.9724/kW h is estimated for a project with profitable life of 25 years with no other financial support.

Keywords: Conventional resources, Solar energy, PV system, Combined Power Generation, Continuous Power Supply

Abstract: Government agencies are using maximum numbers of resources to improving security systems as result of recent terrorist events that dangerously exposed flaws and weaknesses in today’s safety mechanisms. Badge or password- based authentication procedures are too easy to hack. Biometrics represents a valid alternative but they suffer of drawbacks as well. Iris scanning, for example, is very reliable but too intrusive; fingerprints are socially accepted, but not applicable to non- consentient people. On the other hand, face recognition represents a good compromise between what’s socially acceptable and what’s reliable, even when operating under controlled conditions. In last decade, many algorithms based on linear/nonlinear methods, neural networks, wavelets, etc. have been proposed. Nevertheless, Face Recognition Vendor Test 2002 shown that most of these approaches encountered problems in outdoor conditions. This lowered their reliability compared to state of the art biometrics.

Keywords: badge or password, Biometric, fingerprints, socially, face recoginition, neural network,

Abstract: Flooding is one of the most devastating natural events that can occur. The ability to forecast this occurrence has a significant impact on the well-being of humans and other natural beings. According to a brief history of weather study, the ancient Mayans were able to predict floods using planetary motions that were not very accurate. With the advancement of technology and the increasing reliance on computers, people are now able to collect vast amounts of data of various types, such as planetary positions using mathematical models, weather data using rain gauges, and wind turbines. It is very difficult to analyse this data and provide an output, but with the help of a machine learning algorithm, one can gain higher accuracy to forecast floods and inform the region ahead of time, avoiding priceless losses.

Keywords: Gaussian Naïve Bayes, K- Nearest Neighbours, Support Vector machine, Logistic Regression, Decision Tree Classification.

Abstract: Stroke is a medical condition in which the blood vessels in the brain rupture, causing brain damage. If the brain supply of blood and other nutrients is compromised, symptoms could develop. Stroke is the leading cause of death and disability worldwide, according to the World Health Organization (WHO). Early awareness of the numerous stroke warning symptoms can assist to lessen the severity of the stroke. To forecast the likelihood of a stroke happening in the brain, many machine learning (ML) models have been developed. This study uses a variety of physiological characteristics and machine learning methods to train four different models for reliable prediction, including Decision Tree (DT) Classification, Random Forest (RF) Classification, SVM, K-Neighbor classifiers. The datasets downloaded from Kaggle website was used in the development of the approach. The accuracy of the models employed in this study is substantially greater than in earlier studies, showing that the models utilized in this study are more trustworthy. The scheme may be determined from the study analysis, which has been proven by numerous model comparisons.

Keywords: DT, RF, SVM, K-Neighbors, Resnet-34, Vgg-16, Densenet-121.

Abstract: People presently suffer from several of diseases as a result of their lifestyle and the surroundings. As a result, being able to predict illness at an early stage is critical. Doctors, on the other side, find it difficult to make specific estimates based on symptoms. For accurate disease prediction, existing systems use the KNN and CNN machine learning algorithms. Illness symptom collection is essential for disease prediction. Existing system gives lots of time to execute as well as it does not gives accurate results. Because of this we propose our system by using CNN algorithm for better accuracy and to reduce time execution.

Keywords: CNN, KNN, disease prediction, data processing, machine learning.

Abstract: This research examines and assesses methods for detecting false news from four perspectives: the incorrect information it contains, the distribution patterns, and the source's reputation. Based on the review, the survey also identifies some prospective study subjects. We discover and explain fundamental foundational principles in a number of areas to encourage participation. Fake news is the subject of interdisciplinary research. We believe that this survey will help to facilitate collaborative efforts. To propose a solution, experts from the fields of computer and information sciences, social sciences, political science, and the media were brought together. Examine fake news to see if such efforts may improve the accuracy and efficiency of fake news identification. Most importantly, it's straightforward to understand.

Keywords: Social Media, CNN, Machine learning (ML),Deep Fake(DF);

Abstract: Sale energy is an and loss unsoiled an prospective energy an source. Among all other non -conventinal energy option. The sale pv system depends on a geargraphical location on a type of pv modual implemente. PV system are beneficial in areas having a mental amount of incident solar radiation. Solar pawer is used to charge batteries so that solar pawer devices can be used at night. The batteries can obtain be large and heavy taking up space and needing to be a place from time to time. The 1 KW off gride solar pawer plant to full field the ever-grading need customer. More ever use provide solar pawer plant project by our experts and professional. The solar pawer plant design service is non for their reliability and affordable rules. The study is based on design of solar PV system and a case study based on cost analysis of 1.0 KW off grid photovoltaic energy see. The both monthly and weekly cost of energy produce by the one 1 KW pave system having been calculation. The solar 1 KW system can be given internal rate of return of about 1.714% on investment. An annual average solar radiation off about 5.4 Kwahu/m a day. The total amount of energy generated by the system and various losses occurring in the system. 1 KW pave system of rupees 0.9724 .0.9724/kwh is the estimated for a project with profitable life for 25 years with no other financial support. this translates to rupees 80000 payment area the livelier cost of energy of 1kwh generated by the system. 1 kw pave system is also very useful in rural areas of India.

Abstract: Managed Areas For The Production Of Plants Are Greenhouses. Because Current Greenhouse Plants Restrict Themselves, They Are Not Automatically Controlled And Have To Be Manually Operated With Various Documents. The System Suggested Must Be Monitored And Controlled Continuously To Ensure Optimal Growth Of Plants, E.G. Temperature, Moisture, Soil Humidity, Light Intensity Etc. This Work Shows A Management Mechanism For Children's Nurseries Over The Internet Of Things (IOT). The System Can Check For Evident Conditions, Such As Humidity, Soil Immersion, Temperature, Fire Proximity, Strength Of Light, Etc. With Nodemcu Esp 8266, All Data From The Environment Parameters Are Sent To The Nube. If A Parameter Exceeds The Limit Set, The Associated Actuator Is Switched On. If The Earth Parameter Does Not Meet The Required Value, The Microcontroller Turns On The Motor. A Mobile Phone And Desktop Allows The User To Display And Monitor Parameters.

Abstract: Online shopping is popular now a days, but this is popular large retailers. This work aims to concentrates on optimizing the current shopping system in a way that it can be adopted by any small-scale retailers, by making use of an e-commerce mobile application, a website to assist retailers, and a centralized delivery system, using latest technologies like Robotic Process Automation (RPA), Global Positioning System (GPS). This system will provide a great platform for all those shops which has trivial recognition to get exposure and will significantly reduce the requirement of human interaction thus saving a lot of time. Key words: RPA, GPS

Abstract: Security is one of the most important aspects since the dawn of today’s civilization. A smart home indicates an application for different technological implementations, it could indicate any system which controls the door lock and several other devices. Facial identification which is an important section to achieve surveillance and safety, especially for handicapped people, can be considered as one of the ways that deal with biometrics and performed to identify facial images via utilizing fundamental features of the face.

Keywords: Face capturing module, Face recognition module, Voice command recognition module, Buzzer, Raspberry Pi.

Abstract: Pneumonia is a severe pulmonary infection caused by bacteria, viruses, or fungi that infects the lungs and causes inflammation of the air sacs as well as pleural effusion, a condition in which the lung fills with fluid. They are responsible for more than 15 percent of deaths in children under the age of five. It is most common in developing and underdeveloped countries, where overpopulation, pollution, and unsanitary environmental conditions complicate matters and medical resources are limited. Thus, early diagnosis and treatment can play a critical role in preventing the complaint from becoming fatal. Radiography (X-rays) or computed tomography (CT) of the lungs is frequently used for diagnosis. It is frequently utilized for opinion to perform a radiological examination of the lungs using computed tomography (CT), magnetic resonance imaging (MRI), or radiography (X-rays). An inexpensive, non-invasive way to examine the lungs is via X-ray imaging. The pneumonic-white ray's infiltrates (shown with red arrows) separate a pneumonic illness from a healthy one. Casket-ray investigations for the detection of pneumonia are still subject to individual variability. As a result, the detection of pneumonia must be automated.

Abstract: The purpose of the study was to determine the gender differences in academic stressors and reaction to stressors in Physical Education students . In all, 50 male Physical Education students and 35 female Physical Education students during the academic year 2021-22 selected as a sample size for the study. The academic stressors and reaction to stressors measure through the the Student-life Stress Inventory . The result reveals that significant differences were found in academic stressors and Reactions to stressors between male and female Physical Education students. male Physical Education students having high academic stressors and reaction to stressors and female students are more sever stress.

Abstract: This paper proposes an innovative digital subscription model called the dynamic value-based subscription model for companies in the wake of increasing privacy concerns and the imminent deprecation of third-party cookies. I examine how businesses can leverage first-party data to create value-driven subscription offerings while enhancing user privacy and experience. The research investigates the application of machine learning techniques in personalizing subscription tracks and optimizing bundling strategies. By analyzing current trends and future projections, I propose a framework for digital companies to transition from ad-dependent models to privacy-centric subscription-based approaches. My findings suggest that the personalized, data-driven subscription model can not only compensate for the loss of third-party cookie data but also foster stronger customer relationships and sustainable revenue streams in the evolving digital landscape.

Keywords: Digital Subscriptions, Cookies, Machine Learning, Privacy, First-party Data, Third-party Data