VOLUME 10, ISSUE 1, JANUARY 2021

Abstract: In many fields, including agriculture, neural networks have become a very effective method. In this paper, we discuss the applications of neural networks in the field of agriculture, including their advances, specifically in classification, decision-making, pattern recognition, crop yield prediction, plant identification, classification of weed images, remote sensing, identification of plant diseases, precision farming, and agricultural enhancement spatial data analysis. Among these, computer techniques in the field of agriculture are also based on neural networks, especially in the sense of soil and water. The survey was used to convey information about applications, processes, future innovations, and challenges in applying Artificial Neural Network (ANN) techniques in agricultural innovations.

Keywords: Artificial Neural Networks, Agriculture, Soil Classification, Crop Management, Plant Disease

Abstract: Electronic Commerce (E-Commerce) is the buying and selling in products, services using computer networks over Internet. E-Commerce draws on technologies such as mobile, electronic transfer, supply Management, Electronic Data Interchange (EDI), Payment gateways etc. Modern E-commerce typically uses World Wide Web (WWW) for one part of transaction's life cycle. People in the developed world and a growing number of people in the developing world now use e-commerce websites on a daily basis to make their everyday purchases. Still the proliferation of e-commerce in the under-developed world is not that great and there is a lot to desire for.[1]

Keywords: E-Commerce, EDI, WWW, services, Management

Abstract: In computer science, session hijacking, sometimes also known as cookie hijacking is the exploitation of a valid computer session—sometimes also called a session key—to gain unauthorized access to information or services in a computer system. The session established between the user and the server can be hijacked by an attacker by masquerading as an authorized user called Man-in-the-Middle (MITM).The target of the attacker is to have access to users’ confidential records in the server for their own financial gain. In particular, it is used to refer to the theft of a magic cookie used to authenticate a user to a remote server. It has particular relevance to web developers, as the HTTP cookies used to maintain a session on many web sites can be easily stolen by an attacker using an intermediary computer or with access to the saved cookies on the victim's computer (see HTTP cookie theft). Cookie hijacking is commonly used against client authentication on the internet The security of Web applications have been a great concern to many online services. The paper, therefore developed a web application for e-Commerce for the detection and prevention of cookie hijacking in order to protect individual records from unauthorized user.

Keywords: Cookie, Cookie Hijacking, Security, Vulnerability, Authentication, HTTP, Web Application, MITM.

Abstract: A quantum computer has many advantages over a classical computer for exhaustive search. It can tackle problems in science, chemistry, and mathematics that are well beyond the reach of supercomputers. With every additional quantum bit computing power doubles. In this paper I will introduce the basic concepts of quantum computing, and explain the major open challenges in the realization of large-scale quantum circuits systems. It broadens the scope for implementation as it demonstrates quantum mechanical algorithms that can adapt to available technology.

Keywords: Classical computers, quantum computers, quantum computer systems, quantum simulators, Shor’s algorithm

Abstract: Online shopping increasing rapidly Now a day, a huge amount of hindi reviews is present on the E- commerce website. In this paper we have proposed a strategy for classifying given Hindi texts in to different classes and then extract sentiments in terms of positive, negative and neutral for identified classes. Negation is also handled in the proposed system. There are mainly two approaches used for sentiment analysis- lexicon based and machine learning based approach. We emphasis on lexicon based approach which depends on an external dictionary. The system classifies the reviews as positive, negative and neutral and calculate the score for Hindi language. The Methodology used in proposed system is Hybrid approach and Modal is Statical based.

Keywords: Lexicon based approach, Hybrid approach, Statical based model.

Abstract: Bookworm is an app for students.it's basically a notes app modified with feature like do not disturb, e library, manga section with time limit. App is build on android studio with Firebase for user authentication and cloud storage.This app can be run on android with API 29 or above. This can be used by anyone not just for student can be used for entertainment purpose or drawing purpose and much more. .it's user authentication protects your password and keeps your data secure.

Keywords: Firebase,API,android studio

Abstract: Image statistics protection is the necessary segment in communiqué plus multi-media globe. At some point in accumulating and contributing, keep away from 3^rd party entrance of statistics is the confront lone. As long as protection of statistics is the smart job and talent too. A lot of defence procedures are utilized in current duration. Security perhaps given of a statistics is changing the unique in to a number of unidentified structure, indications, draft and so on, which is not recognized by anybody. The finest method of image statistics protection is “Cryptography”. Crypto implies “concealed‟ and graphy implies “writing”. 2 procedures of cryptography: encryption and decryption. Encryption attains the exchange by grouping a key of unique statistics keen on illegible shape known as encoding. Renovating of encrypted statistics into unique shape is known as decoding or decryption. In cryptography key, code or password plays a very important role. This study shows the concert of encoding plus decoding of an image utilizing a solitary key procedure and experienced on a number of images and presents well outcomes.The LSB supported methods are extremely accepted for Steganography in spatial domain. The easiest LSB method purely swaps the LSB in the cover image among the bits from clandestine (secret) data. Additional highly developed methods utilize a number of measures to recognize the pixels wherein LSBs can be swape by means of the bits of clandestine data. The method which is known as DCT:  the placing of clandestine data in carrier depends on the DCT coefficients. Several DCT coefficient assessments over appropriate verge are a possible position for placing of clandestine data.

Keywords: Steganography, Discrete Cosine Transform, Least Significant Bit.

Abstract: In this universe of developing advancements everybody is utilizing Android. With huge number of work opening the human labor force has increased. Thus there is a need of a framework which can deal with the information of such countless Employees in an association. This undertaking disentangles the assignment of keep up record in view of its easy to use nature. The Aim of "Employee Task Management System" is planning a booking framework for a work community. Workers are the foundation of any organization. The executives of worker execution assume’s a significant part in choosing the accomplishment of the association. Worker the board application is an incredible asset to calm the client from the convoluted errand of dealing with representative planning physically. Representative administration application utilizing android PDA's is attempting to fabricate an android application for log the information on the worker naturally. The application is really a set-up of uses created utilizing Android and PHP. It is easy to comprehend and can be utilized by any individual who isn't even acquainted with straightforward representative's framework. It is quick and can perform numerous activities of an organization or association. The product is very easy to understand. The task contains modules like Employee and Admin. This form of the product has multi-client approach. For additional upgrade or advancement of the bundle, client's.

Keywords: employee management system, employees, human resources, leave management, task management, Android, PHP.

Abstract: The project aims to design a dynamic traffic signal system based on density, where the timing of the signal will adjust automatically when the traffic density is detected at any junction. In most cities around the world, traffic congestion is a serious problem and it is therefore time to change more manual or fixed timer mode to an automated decision-making system and also to provide special schedule for emergency vehicle. The current traffic signalling system is based on fixed time, which may make it inefficient if one lane is operational than the other lanes. In addition, the proposed system has innovative services that allow drivers to remotely monitor the traffic rate and the number of parking spaces accessible to their destination using an Android smartphone app to prevent traffic jams and take another alternative route to avoid getting stuck and also to make it easier for drivers to avoid unnecessary trips while searching for a free parking space. In order to connect people to a smart city, our system combines three linked smart subsystems (cross-road management, parking space management, and a mobile application). It also implements the proposed Spherical Routing Protocol (SpRP) and tests its performance metrics.

Keywords: traffic signal, Arduino, microcontroller, automated system.

Abstract: A proxy server is a computer system used for connecting client and server. It is used to create a firewall. It is a gateway between users and the internet. A proxy server is always an intermediary” between end-users and internet that are the web pages user visits online.The word proxy refers as "to act like same on behalf of some other ," means it acts on behalf of the user.

Keywords: VPN, Relocation.

Abstract: Path planning is an essential part of autonomous robots so that they can work in real environments with several obstacles. There are two method of path planning classical and heuristic methods which are categorized according to their efficiency and complexity. Heuristic methods are based on machine learning algorithms which can find more feasible solutions. In this paper a review of heuristic algorithms used for path planning are discussed according to their feasibility and complexity. Different papers on heuristic algorithms are discussed to find out the computational complexity and efficiency of the algorithm.

Keywords: Path Planning; Neural Networks; Fuzzy Logic Based Algorithms; Hybrid Algorithms; Nature Inspired Algorithms.

Abstract: Path planning is the first task for a Robot to autonomously navigate, especially for autonomous Robots. For multi-Robot systems the process is more complex than a single Robot system. The commonly known algorithms for path planning usually finds solutions for single Robot systems and do not propose ideas for multi-Robot systems. In this paper an enhanced Dijkstra’s algorithm for multi-Robot systems with a multi-layer dictionary is used to navigate multiple Robots on an indoor map autonomously and simultaneously. Simulation and Experimental results show that the proposed enhanced algorithm was able to generate paths for the multiple Robots that where navigating through the map simultaneously and assigning optimal or feasible paths for the Robots to navigate through.

Keywords: Dijkstra’s algorithm; multi-Robot systems; multi-layer dictionaries; Path Planning.

Abstract: The Witel Business Service Division is a division that is responsible for the operational sector of PT Telkom Telekomunikasi Indonesia. One of the operational areas in the Witel division is customer service. The task of contacting customer needs in the ICT field as well as the disruption that the customer is experiencing is the task of the marketing department. The marketing department visits or prospects to customers through visits, calls, emails, to each predetermined customer data. If the customer agrees with what marketing offers, a memorandum of understanding is made. Furthermore, marketing will report its activities to the manager. Manager has one of his duties to evaluate marketing performance. There are several parameters / criteria for evaluating marketing performance based on reports received. In this study, to assess marketing performance using the Fuzzy Inference System (FIS) Sugeno method, namely the process of making a mapping system from the input given to the output using fuzzy logic. Fuzzy variables are obtained from the criteria for Existing Customer Management, Acquisition of Subscription Contracts and PBC is Acquisition of Billing Complete which is then used as fuzzy input. The results of this study resulted in the Sugeno method of Fuzzy Interference System which can be applied at the calculation stage of the marketing performance assessment, where the value of each input variable is subject to fuzzification first. Furthermore, inference is made to the rules used and ends with the defuzzification stage in the form of calculating the score using the weighted average method. With the average weight value is in the range 41-89, and is in the good category in general, although there is a low value.

Keywords: Fuzzy Inference System (FIS), Fuzzy Logic, Sugeno Method

Abstract: Cyber security has become an increasingly important area in computer science in response to the expansion of private sensitive information. Intrusion can be defined as an uncertified access, which aims to compromise integrity, confidentiality and availability of data. Conventional intrusion prevention method such as access control firewalls and encryption cannot fully prevent system from advanced attacks. Intrusion Detection System has become a crucial part of computer security, which is used in detecting the above-mentioned threat.This paper presents an agent based Anomaly intrusion detection and prevention system using Reinforcement Learning Technique. The system uses two agents, the first agent attacks the network system while the second agent detects the attack and classify it to be either normal, dos, probe, u2l and u2r attack, the orange line represents the reward receive by the attacking agent while the blue line represents the reward of the agent detecting and classifying the attack. The attacking agent receives a total reward 5 while the defending agent received a total reward of 95. This means that the defending agents performs more better in detecting and classifying attacks that is being carried out by attacking agent. The diagram also shows the loss values of the both agent during training. The both agent has a loss value below 0.5 during training. Figure 5 shows the performance of the defending agent in classifying an attack currently. The agent obtained individual accuracy in each of the attack. The accuracy are as follows, normal 0.79%, DoS 0.94%, R2L 088%, Probe 0.94% and U2R 0.99%.

Keywords: Reinforcement Learning, Deep Q-learning Network, Intrusion detection, Anomaly attack.

Abstract: Smart security system will protect your house,valuables and to keep your family safe. A smart doormat and a smart door is developed to ensure a Smart home.The doormat containing sensors.The camera which is placed in the door will capture the image and send this image as a message to the user.The door will open by scanning face. The face recognition system will store the faces of all the members in the family.If the person is not a member then the door will not be opened and an alert message will send to the users phone that message will contain an OTP number.After entering the OTP the door will be opened .In case if a thief is trying to break the door or trying to enter into the house then a doormat containing sensors is provided along with this.When a person steps into the mat an alert message will be sent to the owner and make an alarm.

Keywords: IoT, PIR Sensor, Ultrasonic Sensor, Local Binary pattern(LBP),Rasbperry pi 4

Abstract: The project deals with day-to-day account settlement. It is a simple application used to make expense management quicker and easier. It uses OCR methodology to read and process receipts which can be converted to spread sheets, documents, or CSV. The user can take/upload a receipt image for recognition of the images. The bill images are captured through camera by image processing and then the data are extracted and also analysed by recognizing the text. All these actions are performed by using an OCR tool, which performs text detection and recognition on the pre-processed image using the Deep Learning (Two-Step CNN Framework) model. Once the image is clicked, it is then encoded and sent to the server. The extracted text data is used to extract the relevant information (such as date, company name, items, total, etc.). This data is then sent back in the desired format

Keywords: OCR, CNN, Text Recognition, Character Segmentation

Abstract: Virtual health care management system plays a critical role in this era where people lead a busy life and does not have enough time to take good care of their health. Eventhough such systems already came into practice where people can monitor their health from home itself, there are many drawbacks to the existing system. A health care system without any basic checkups is of no use. So here we aims at implementing an efficient real time health monitoring system which includes pulse rate checking in addition to doctor appointment so that the user get access to their basic test results from anywhere. In remote areas, it is hard to find laboratories and pharmacies, so this system also provides accurate location of the nearest laboratories and pharmacies.

Keywords: AI, Health monitoring, Pulse rate, Smart phone

Abstract: Acquiring sensitive info from the user in some malicious sites that looks like the legitimate webpage and that they do a sort of criminal activity that's referred to as phishing within the electronic world. associate degree assailant will use this type of phishing or fraud by using such websites, that could be a severe risk to internet users for his or her personal and confidential information. So, within the field of e-banking and e-commerce, this act makes a threat for all webpage users. during this paper in the main discerning the various options of legitimate, suspicious and phishing websites. These options are fed to the machine learning algorithms that are constitutional hence are used for comparison and to ascertain the accuracy of the algorithmic rule. Algorithms utilized in this comparison are J48, Naïve Bayes, random forest and supply Model Tree (LMT) are used and them accurately to predict the web site legitimacy is calculated. Also, the most effective algorithmic rule among completely different algorithms will be selected. during this paper, we'll compare the ends up in the 2 ways that. Firstly, we discover the best algorithmic rule by mistreatment the comparison of the various attributes like properly Classified Instances, Incorrectly Classified Instances, Mean absolute error and letter of the alphabet statistics. Secondly, the accuracy of those algorithms can analyse with completely different parameters like TP Rate, FP Rate, Precision, Recall, F-Measure, MCC, mythical creature space and People's Republic of China space that's visualized within the chart. the chosen algorithmic rule makes the web site analysing method automated. Before creating payment on any e-commerce web site, this prediction model can be used for determinative the legitimacy of that web site.

Abstract: Diabetic Retinopathy (DR) affects 1 in 3 peoples with diabetes and remains the leading cause of blindness in working-aged adults. Currently, detecting DR is a manual processing and time consuming that needs a trained clinician and evaluate digital color fundus photographs of the retina of eye. By the time readers can submit their reviews, often a day or two later, the delayed results lead to lost follow up, miscommunication, and delayed treatment. The need for a comprehensive and automated technique of DR screening has long been accepted, and previous efforts have made good progress using image classification, pattern recognition, and machine learning. This system uses ensemble learning technique to detect diabetic retinopathy from scan images. Inception V-4, Xcpetion, ResNeXt are the three base models used which are modified and optimized to avoid overfitting, underfitting issues. The base models are individually optimized and their predictions from final layers are combined using algebraic combiner (maximum rule). Each base model differ in performance when using different datasets, therefore final output does not always depend on one higher performing base model.

Keywords: Xception, Inception V-4, ResNeXt, Ensemble Learning

Abstract: In this hi-tech world the main problem that visually impaired people facing is social restrictiveness. They suffer a lot in unknown surroundings without any aid. Visual information is most basic information for some of the tasks, in this area visually impaired people experiencing some difficulties that necessary information is not available for them. With the advanced technology it is possible to give a support for visually impaired people and make the daily tasks easier than before. This project is using Artificial Intelligence, Machine Learning, Image and Text Recognition. The idea is implemented through mobile app that focuses on voice assistant, face recognition, obstacle avoidance for path planning. Here we propose a voice operated IPA which can process direct commands to perform menial tasks for the users. By using artificial intelligent and image processing, this smart device is able to detect faces and obstacles. The detection process is manifested by notifying the visually impaired person through a voice notification. Text recognition help those people to read environmental messages, words, letters, daily newspapers, and so on to cope up with the social life. Images will be processed in a wearable smart device, it means, in a light and small processing device. It will be an efficient way to utilize the new facilities and technologies to help visually impaired people to interact with the environment without any external aid.

Keywords: Text Recognition, Image Processing, Face Recognition, IPA, Obstacle Avoidance.

Abstract: Automatic control of home appliances must be artificially intelligent systems that need to make itself depend on human being's action and surroundings. These systems are thoroughly examine user requirements and conditions of the neighbouring in order to estimate future actions and reduce user interactions. Here we bring a new Home automatic control system based on machine learning. In this system we put forward VGG16+LSTM Architecture to increases accuracy and efficiency of the system. The system contains three modes of operations to control home appliances, they are: emotion detection mode, automatic mode and manual mode. This system prominence given to energy consumption and prediction performed by logistic regression and also determine frequency occurring set of devices by using apriori Algorithm .The another component of this system is that we can examine power consumption of individual devices.

Keywords: Machine Learning, Association rules, VGG16, LSTM, Logistic Regression, Apriori algorithm

Abstract: Agriculture have an important role in our day today life. Now adays the population has increased and it also increased the demand for food. So it is necessary to increase the production of crops. For better production of crops the plants should be properly fertilized and protected from diseases. Manual identification of disease is not possible for large scale farmers all the time as it is time consuming and need more labours for that. Diseases can be easily detected by using machine learning. By using machine learning technique we can identify the disease that is affected to the leaf, stem, root and fruits. Images captured using mobile camera will be processed using image processing technique. The features are extracted from the disease affected part and it can be classified using machine learning technique such as Convolutional Neural Network(CNN) and MobileNet. An application that can predict the disease affected to the plant and possible precautions to avoid the disease will be developed.

Keywords: Machine Learning, CNN, MobileNet, Image processing

Abstract: In our nation, fruit recognition and its maturity monitoring is a difficult task due to the mass production of fruit products. In order to determine and evaluate the quality of the fruit accurately. The project presents, In real time with camera developing a sorting machine for classification of multiple fruits and checking the time they survive and evaluate the rank of the fruits based on its quality. Firstly, the algorithm bring out the RGB image values and the background of image was disengaged by the split-and-merge algorithm. Secondly, the extracted multiple features are namely its color, statistical, textural and geometrical feature. Geometrical features are used in the evaluation of quality of the fruits. Additionally four different classifiers k-nearest neighbour(k-NN), support vector machine(SVM), sparse representative classifier(SRC) and artificial neural network(ANN) are used to classify the fruits. The SVM classifier has seen to be more effective in quality evaluation. Using k-fold cross-validation techniques validates the system performance by considering different values of k. The classification is among Rank1, Rank2 and defected one. The system achieved maximum accuracy for fruit detection and classification.

Keywords: Classification, Defect detection, Geometrical feature, Textural feature, Statistical feature

Abstract: More advanced technologies in the field of automobles had come up recently. The main focus is on making the vehicles automated. There are various aspects in the field of automobile industry which makes the vehicle automated. This autonomous vehicle is based on the idea of using fewer resources and less cost to develop a highly efficient self-driving vehicle. It is a mix of Robotics, Machine learning, and AI and using an ARM-based low-power Low-cost CPU (P.C) which is mainly found on Mobile architecture by replacing a CPU of X86 architecture which is having high cost and needed high power consumption. The vehicle is trained using Machine Learning and it detects not only simple obstacles but also understands several kinds of signs, stopped vehicle, take a turn when a parked vehicle or comes in front of it, understands the road line and follow it, only by using a simple camera and a low power low-cost simple CPU. When an obstacle/sign captured by the camera, vehicle takes a needed decision by sending the data into the Machine learning algorithm in PC. Thus, the vehicle will be trained enough to navigate safely by relaxing the driver.

Keywords: IoT, Robotics, Autonomous Vehicle, Machine Learning

Abstract: The Internet of Things (Iot) is able to connect Millions of devices and services at any time, from anywhere in the world with the help of Internet. It also plays a major role in filling the gap between all the day to day things and the networking system, and thus creates a way to access all the non-internet objects from any distant location. With the growth of population and commercial development increasing day-by-day, the number of vehicles on the road is also increasing day-by-day. So that, spending too much time for searching a parking area/slot in the city will be a time waste process, which will lead to greater financial costs. Thus, it is very much important to develop an automated smart parking management system that would help the user to find a perfect parking space without wasting any time. In the proposed system, a user can search parking areas nearby the user’s location, choose and book for a slot in a parking area from the smartphone. The system also comes with real-time detection of improper parking and automatic parking payment collection. The proposed system in future will surely help users to overcome the difficulty of parking and also saves much of the user’s time.

Keywords: IoT, Cloud Computing, RFID Technology, Ultrasonic sensors

Abstract: Alzheimer’s disease is one among the most common cause of dementia among older adults. Dementia is the lack of cognitive functioning thinking, remembering, and reasoning, and behavioural skills to such an extent that it interferes with a person’s everyday lifestyles and activities. Dementia levels in severity varies from the mildest stage, while it's miles just starting to have an effect on a person’s functioning, to the maximum severe degree, while the character ought to depend absolutely on others for basic activities of day by day living. Alzheimer’s disease is currently ranked as the 6th leading cause of demise in the united states, but recent estimates imply that the disorder can also rank third, simply at the back of coronary heart disorder and cancer, as a cause of demise for older human beings. The causes of dementia can vary, relying on the sorts of mind adjustments that may be taking vicinity. Different dementias encompass Lewy body dementia, frontotemporal disorders, and vascular dementia. It is not unusual for human beings to have blended dementia. A mixture of two or extra sorts of dementia. An early Alzheimer’s prognosis provides you with a better danger of benefiting from treatment. An early diagnosis makes individuals eligible for a greater diversity of clinical trials, which strengthen research and might provide medical blessings. An early diagnosis opens the door to future care and remedy. It enables people to plot beforehand while they're still able to make essential selections on their care and guide desires and on financial and legal topics.

Keywords: Inception v4, Deep Learning, Artificial Intelligence, Alzheimer Disease, Magnetic Resonance Imaging, Image Processing, Convolutional Neural Networks

Abstract: In Large strides being created in digital technology that digital design hasn't wedged effectively. Our application could be a step in this direction, permitting users to look at a 3D rendered model - a virtual likeness of the physical furnishings with no interruption of the markers - which may be viewed and designed in period victimization of our AR application. This study proposes a replacement methodology for applying increased Reality technology to interior style work, wherever a user will read virtual {furniture|piece of furnishings|article of furniture|furnishings} and communicate with 3D virtual furniture information employing a dynamic and versatile computer program. A.

Keywords: AR, 3D Rendering, K-means Algorithm

Abstract: Nowadays Musculoskeletal disorder is increasing day by day in humans because of accidents, and because of the busy schedule of employees, they also face an identical problem and also the people between age of 35 to 60 facing the same problem. to beat this problem physical exercises can reduce this disorder and yoga could be the best choice for this, yoga is an important physical exercise for all the issues.in short, yoga is the best medium. but for doing yoga we'd like proper training and also a trainer who will monitor accuracy, body movement. so, to boost the popularity accuracy with reduced training times, we must be prepared Microsoft Kinect to recognize different mutual points of the human body in real-time and from this mutual duration we assume various angles to live the accuracy of specific yoga poses for a user. To test also, explain the recommended program, we decided video series of yoga. Our proposed system can successfully recognize different yoga poses in real-time.

Abstract: Here a Brain Computer Interface (BCI) technology to analyze the brain waves along with controlling a device as well as analysis of the EEG signals BCIs are interfaced through the humanoid brain signals along with gadgets by decoding brain activity in real-time action commands. The project will be implemented using ML [machine learning] and IoT[Internet Of Things].

Keywords: EEG, IOT, BCI, IoT, ML, EPOC.

Abstract: Recent scanning advances have enabled a variety of applications across all walks of life. One place to apply is the separation of new products .This paper shows the latest improvements in scanning the application. These applications are used in normal life. Many fruit scanning applications to scan the amount of fruit ex. vitamins, proteins etc. For the past 15 years, hyper spectral imaging has emerged as a new generation of food quality sensor technology and safety tests, as it incorporates major imaging and spectroscopy features, enabling both visual and location information to detect an object simultaneously. This app helps speed up the process to improve accuracy and efficiency and reduce time. This photo collection from camera and gallery to show fruit name and brief description for user. This app incorporates three analytical methods: color-based, shape-based and size-based.

Keywords: fruit, scanner, pattern recognition, analysis, data base.

Abstract: Ride-sharing is a service that enable drivers to share trips with other riders, contributing to appealing benefits of shared travel cost and reducing traffic congestion. Most current ride-sharing system, however, depend on a central third party to organizing the service, subjecting them to a single point of failure and privacy concerns about disclosure by both internal and external attacks. The proposed system makes without depending on a trusted third party, drivers to provide ride-sharing services. Both riders and drivers can learn if they can share rides while maintaining their travel information, including place of pick-up/drop-off, departure/arrival date and price of travel. However, Malicious consumers to send multiple ride requests or requests, exploit the anonymity given by the public blockchain In order to find a better deal or to make the offer, while not committing to either of them, offers unreliable structures. The proposed system addresses this issues by implementing a time-locked method. A ride-sharing deposit protocol by leveraging the smart contract and zero-knowledge collection evidence for membership. In a nutshell, a driver and a passenger will have to demonstrate their good will and commitment to the blockchain by submitting a deposit. Later, a driver must prove himself to the blockchain on the decided pick-up time that he/she arrived at the pick-up place on time. To protect the privacy of the rider/driver by hiding the exact pick-up spot, the evidence is performed using evidence of membership of the zero-knowledge set. To ensure equal payment, moreover, A pay-as-you drive methodology is applied depending on the driver’s elapsed time of the rider and driver. Furthermore, we implement a model of reputation to rate drivers based on their history behaviour without any third party interference.

Keywords: Blockchain Technology, Decentralization, Consensus Algorithm, Distributed System, Cryptocurrency

Abstract: As we know, the advancements in the Radio Frequency and Messaging technologies have made a platform to come up with various innovations reducing human effort. Since online shopping has become a part and parcel of common man’s life, this is the right time to make use of existing technologies to simplify the procedure. The basic idea of the work is to introduce technology into our lives for monitoring issues which demand our personal presence. The aim is to provide a reliable and user friendly solution to problems incurred during online shopping. A stand alone box is designed which receives and stores the intended parcel so that the customer can retrieve it as and when required.

Keywords: SMART Locker, IOT network, Cloud, Sensors

Abstract: Smart mirror is a device where would be able to see news, temperature, weather and can also schedule events for one month. We can interact with smart mirror using voice command, hand gesture and smart phone. Our system uses raspberry pi based processor board along with display and IoT based circuitry and temperature sensor. A camera used to capture the gestures and gives corresponding output according to the programming . The Internet of Things allows devices to communicate with each other in different and important places at the same time. One of the most important IoT applications is the smart mirror. It is a mirror that acts as a reflective surface and as an interactive screen at the same time. It provides valuable information on the display at a glance, while also acting as a conventional mirror. Smart Mirror is a mirror which allows touch-free user interaction with important information displays such as current news, time, date weather, schedule, Temperature and setting up of reminders in the form of widgets on the screen, while also providing notifications or alerts to the smart phone using an application. It uses innovative technology to achieve an interactive system, made for ease of access to basic important information and enhancing utility.

Keywords: Smart mirror, Hand gesture, Face detection, Voice command, Hand detection

Abstract: Having a lacklustre attendance system or process for employees creates inconsistency in their daily schedule. Employees coming to work a few minutes before the login time may face traffic at the attendance device causing incorrect data records to be registered although they were there well in advance. Organizations have to set up these devices all around the campus making it an expensive investment that may need further servicing throughout its use. Faulty devices will hinder employee login unless immediately fixed. This results in the need for an intelligent process to provide a simplified and flawless daily login for the employees. The absence of a technique capable of dealing with the above-stated complications led to the development of this innovation. This paper presents a novel and efficient technique integrating a smartphone application to simplify biometric login through the fingerprint scanner and a web application for information and a subscription-based package for users. A database stores the user information securely for verification purposes.

Keywords: Android Studio, Smartphone, Fingerprint Scanner, React Js.

Abstract: Education institute are the nursery for the future minds of the nation. Knowledge represents the intangible assets of the education institutes, industries and nations. With development of the information and telecommunication technology activities like commerce, communication, entertainment and learning are occurring on internet. As universities and colleges are started using an online learning platform for providing content to students and started using student management system for better management of the students personal data. Education institute have large amount of student data like basic personal information, attendance, marks, achievements etc. Online learning platform provide an opportunity to capture fine gain data about student online activities like course content he browse, time spent on each unit, post on forum, practice test, sequence of activity will generate large amount of structured and unstructured data. But it is found that educational system are notoriously poor in managing the data and taking advantage of this generated data. There are two research area for Big Data mining in education called educational data mining and learning analytics. Educational data mining is suit for the computational and psychological methods and research approach for the understanding how student learn, predict student future learning behaviour. Learning analytics is becoming defined as an area of research and application and is related to academic analytics actions and predication analytics. Recently waste amount of work has been done in other area like ecommerce portal and increase the click through rate. So now considering opportunities with data generate by the online learning platform we can mine the education data for calculating learner performance interest problem face by the learner knowledge level for different knowledge point. As recently there are lot of research in education data mining and some researcher started treating data in education system as big data problem, we have done survey of various research in education.

Keywords: —Education, Data Mining, Predictive analysis, Big Data, Hadoop, Association Rules.

Abstract: Handwriting recognition is the ability of a machine to receive and interpret handwritten input from multiple sources like paper documents, photographs, touch screen devices etc. Recognition of handwritten and machine characters is an emerging area of research and finds extensive applications in banks, offices and industries. The main aim of this seminar is to design expert system for, “Handwritten Character Recognition using Neural Network”, that can effectively recognize a particular character of type format using the Artificial Neural Network approach. Neural computing Is comparatively new field, and design components are therefore less well specified than those of other architectures. Neural computers implement data parallelism. Neural computers are operated in way which is completely different from the operation of normal computers. Neural computers are trained (not Programmed) so that given a certain starting state (data input); they either classify the input data into one of the number of classes or cause the original data to evolve in such a way that a certain desirable property is optimized.

Keywords: Artificial Neural Network; Handwritten Recognition; Image Processing; Feature Extraction.

Abstract: Feature selection is a big task and its challenge for high dimensional data. Semi-supervised feature selection is a combination of supervised and unsupervised data. Supervised data means labelled data and unsupervised data means unlabelled data. In semi-supervised feature selection unlabelled data is more than labelled data. Supervised learning means well labelled data that means data is already labelled with correct answer. Unsupervised data means that data is neither labelled nor classified and allowing algorithm to process without guidance. Supervised learning use processes like regression and classification. Algorithms used for unsupervised learning clustering and association. In supervised learning optimized performance criteria with the help of previous experience.

Keywords: Supervised, Semi-supervised, labelled, and unlabelled.

Abstract: The Internet by far plays a major role in people’s life. It has drastically improved the quality of life and the standard of living of so many people. It has widened its branches into many different levels and areas. The ecommerce industry is one such branch which has come into spotlight in the recent years. The online bookstore system has eased the life of so many book lovers by making it easy for them to purchase books online. It is not always feasible to access a traditional bookstore, it is limited by its operation time, availability of a particular book, its location and most importantly its capacity and the space required to store numerous books. Such drawbacks have led to the evolution of e-commerce industries related to bookstores. Our project is one such simple e-commerce website which houses various books of different categories for a consumer to purchase online.

Keywords: Internet, e-commerce, traditional bookstore, online bookstore, website.

Abstract: Nitrogen vacancy (NV) centre is a small imperfection in the structure of diamond. These NV centres due to their unique properties find applications in various fields like magnetometry, Nano-MRI etc. This is why it is considered to be one of the most promising new technologies for the future of nanoscale magnetic measurements and imaging. N-V colour centres in a diamond lattice are of utmost significance due to their photoluminescence properties. It can achieve significantly high degree of spin polarization with optical illumination. In addition to the defect’s quantum spin states interacting with magnetic fields, the spin state of N-V centres is especially long lived and provides a long coherence time to allow magnetometry with the ability to remain intact up to a few milliseconds at room temperature. This paper throws light on various concepts related to Nuclear Quadrupole resonance (NQR) with respect to NV colour centres in the diamond lattice. In future, the use of these individual, atomic-sized quantum systems can thereby be used inspire new concepts in the fields of cell biology and solid-state physics too.

Keywords: Nitrogen vacancy, magnetometry, photoluminescence, quantum spin, Nuclear Quadrupole resonance.

Abstract: Portable technologies offer new opportunities for learning, in which, learning collaboratively has become an extremely important skill. Technology causes learners to be more engaged and often retain more information. Funding for technology has brought teacher and learner digital devices to a large number of classrooms in basic education. Given the vast resources invested in digital devices by the governments, measuring the level of ICT integration for collaborative learning in basic education is of crucial importance. The study established that 57.3% of variation in Collaborative learning in Basic Education was accounted for by Indicators of ICT Integration, The ICT integration framework for collaborative learning (IIFCL) so developed showed the critical indicators to measure the level of ICT integration for collaborative learning in basic education.

Keywords: Collaborative learning, ICT Integration, Portable technologies, Digital devices.

Abstract: Using Artificial Intelligence and Machine Learning our project proposes a system named Dictation Module. It is a device that can help society to ace their work and help people in multitasking. As it is rightly said that “If a person cannot perform multiple tasks at the same time, he cannot achieve everything he wants”. Multitasking is the basic asset of every individual in the society. Unlike others, the socially active businessmen or lawyers are forced to be dependent on this aspect. So we have proposed a device that helps to do multiple tasks anywhere you want. Imagine you are driving and you realize you forgot to mail someone something very important, how will you do it in the traditional way is to reach out to the destination and then start typing your mail and send it, which will cost you lots of time as well as some mishap can occur due to this process. But using our module you can dictate the mail to the machine and tell it to send it to the desired person while driving the car simultaneously. Also Dictation Module can be proven very useful to visually impaired or physically debilitated people.

Keywords: Artificial Intelligence, Machine Learning, Voice Recognition, Speech Recognition, Speech To Text conversion

Abstract: This work presents the review on concept of image fusion method by using DNN. The CNN and DBN system has a problem with accuracy, then proposed method is used for improving the system. The fundamental test of picture de-obscuring is to devise effective and dependable calculations for recuperating however much data as could reasonably be expected from the given information. The use of DBN network in existing system works only to reduce error in system. Due to this, it requires better deep learning method for improving accuracy of system. The CNN method uses only 2 convolutional layers for feature mapping. But the proposed method uses 5 convolutional layers and 3 overlapping layers. Due to this, it will help to improve accuracy of system as compared to other existing methods.

Keywords: Image Classification, Machine Learning, Deep Learning, MATLAB etc

Abstract: Search engines and web crawlers aim to provide users with the most relevant and useful information in response to their search queries.In web crawling, the relevance of a page refers to how well it matches a user's search query or a specific topic. Search engines uses various complex algorithms to determine the web page relevance to ensure that users receive the most accurate and useful results when they perform a search.