VOLUME 11, ISSUE 10, OCTOBER 2022

Abstract: Using the special ability of Generative Adversarial Networks (GANs) to create fresh adversarial instances for model retraining, we offer a novel defense strategy against adversarial examples in this study. In order to achieve this, we create an automated pipeline that combines a convolutional neural network that has already been trained with an external GAN called the Pix2Pix conditional GAN. This pipeline allows us to identify the transformations between adversarial examples and clean data as well as create new adversarial examples on the fly. In an iterative pipeline, these adversarial samples are used to strengthen the model, attack, and defense. Our simulation findings show that the suggested strategy works well.

Keywords: adversarial machine learning; botnet detection; generative adversarial networks; machine learning

Abstract: Semantic segmentation is needed by regional planners to know the composition of land cover in their area, so that they can take the right policy. Several methods from manual to automatic have been researched, both based on colour and pattern. Each method has their strong and weakness, so it is necessary to make the right choice when applying the method. Currently, multispectral imagery is still very rarely used, even though sources of information from the internet are easy to find, i.e. Landsat imagery from the United States Geological Survey. This study uses two methods for segmenting three-channel multispectral images (red, green, and blue), namely iterative self-organizing clustering (ISOCLSUT), which is based on a colour sensor, and a multiresolution algorithm, which is based on colour and pattern. For the experiment, the pre-processed satellite image of Karawang district was segmented using the ISOCLUST as well as multiresolution algorithm. The experimental results show that land cover segmentation with multiresolution algorithm is better than ISOCLUST for RGB but for more than three channels, i.e., seven frequency channels, ISOCLUST shows better performance compared to real image conditions.

Keywords: Satellite Imagery, Multispectral, ISOCLUST, Multiresolution

Abstract: The basic principle of this algorithm is to select the optimum raster location to represent a straight line. To accomplish this algorithm always increments either x or y one unit depending on the slope of line. The increment in the other variable is determined by examining the distance between the actual line location and the nearest pixel. This distance is called decision variable or error.

Keywords: Pixel, Decision Variable, Error.

Abstract: This papert deals with an E-health web application framework, cloud platform and responsive web design which aim to adjust the presentation on mobile devices. This work presents the whole development process of the self-care management web-app framework which provides instructive support for future other E-health field applications. The report consists of the following main parts: analysis, design and implementation, and evaluation. Literature review and internet search are the main methods for making an investigation of existing systems and related works. A prototype is developed by using .Net, CSS3, Javascript, and HTML5 technologies. The system test and evaluation are made to show the system’s usability. The Interoperability of E-health model system allows people store large amount of information in different place. In many of the developed countries, healthcare has evolved to a point where patients can have many different providers– including primary care physicians, specialists, therapists, and even alternative medicine practitioners – to service their diverse medical needs. Telemedicine system is receiving great importance due to current changes in healthcare sectors all over the globe. The need for medical sectors to provide appropriate and precise remedy for various diseases is essentially increasing as they are facing new challenges every day. There comes a big problem that the information sharing increased the risk of medical misuse and data theft. The E-health record may include the patient personal information, like telephone number, age and so on, even more, the diabetes patients’ glucose, exercise information which are private, sometimes, the patients just want to share their relative information to their physician. Data theft can invade to patients’ medical records and stole patients records to do financial fraud. In order to forbidden this crime, how to keep the privacy and security becomes the key point in our work.

Keywords: Telemedicine, Data-Mining, Tools, Techniques, Medical-Data.

Abstract: The world today is governed by automation, In sd most of the domain, automation has maneuverer industrial advances and has become predictable. But in a lesser degree, the contribution of automation to agriculture. When complex operations are made automated to simplify tasks, the benefits of automation can also be tapped to perform simple household tasks also. Cultural irrigation is made cost-effective by considering soil, temperature and humidity levels of moisture. Health supervision is performed by an autonomous agricultural rover that moves around the field and collects data through a camera attached to it. Using the Ultraviolet Flame sensor, it is identified in case of fire incidents and the fire is placed off. The farm also uses a PIR sensor and buzzer to protect it from animal intrusion. SWEEPY , the smart floor cleaner is both an autonomous and manual controlled cleaning machine used to simplify and achieve the task of cleaning. By means of its dry and wet modes all round cleanliness and hence good health is achieved. Index Terms - IOT, Image Processing, Machine Learning, Open CV, Laptop, IP Camera.

Abstract: The use of social media is on the rise right now. On travel websites, millions of users evaluate and rate tourist destinations every day. These reviews may be subjected to sentiment analysis, which will be useful in determining the popularity of tourism destinations. Tourists are able to choose their tour destination with ease based on the results of the sentiment analysis. A use case is a common description of an entire transaction including several procedure objects. Utilizing use case diagrams, such as sequence and cooperation diagrams, the UML language provides an appropriate framework for scenario acquisition. In this paper the various UML Diagrams as a part of the design of the sentiment analysis system is shown The Dataset has been collected from various tourism review websites. Keywords: UML, Sentiment analysis, machine learning, tourist place

Abstract: This document gives formatting instructions for authors preparing papers for publication in the Proceedings of an International Journal.  The authors must follow the instructions given in the document for the papers to be published.  You can use this document as both an instruction set and as a template into which you can type your own text.

Keywords: nodes, address, data, link, head.

Abstract: In the fourth industrial revolution, all industries are installing sensor-based equipment for better manageability and performance. All these sensors continuously transmit the data and are monitored mainly by neural network-based smart systems. Due to a lack of proper knowledge, utilizing all available resources to maintain the complexity of the operation in securing the control systems is very challenging. Making decisions using traditional technology and software is more difficult to safeguard information against security threats successfully. This paper provides an overview of the artificial intelligence and cyber security implementation prospects in securing IoT-based control systems in the heavy equipment industry. Securing essential information is a concern in today’s organizations. Also, this research gave an important reason for securing Neural Network (NN) information in an organization and the benefits and methods used. The study recommends and concludes steps the organization should implement for securing the IoT and NN-based devices also indicate that the organization should contain backup for retrieving information based on the organization's competency.

Keywords: Industry 4.0, IoT, sensors, heavy equipment, AI, NN, Cybersecurity

Abstract: The proliferation of user-generated content on social media has made opinion mining an arduous job. As a micro-blogging platform, Twitter is being used to collect views about products, trends, and politics. Sentiment analysis is a technique used to analyse the attitude, emotions, and opinions of different people towards anything, and it can be carried out on tweets to analyse public opinion on news, policies, social movements, and personalities. By employing Machine Learning models, opinion mining can be performed without reading tweets manually. Their results could assist governments and businesses in rolling out policies, products, and events. Seven Machine Learning models are implemented for emotion recognition by classifying tweets as happy or unhappy. With an in-depth comparative performance analysis, it was observed that the proposed voting classifier (LR-SGD) with TF-IDF produces the most optimal result with 79% accuracy and 81% F1 score. To further validate the stability of the proposed approach on two more datasets, one binary, and another multi-class dataset, and achieved robust results.

Abstract: Every year, over 900,000 individuals worldwide are diagnosed with Leukemia, i.e., Blood Cancer, but many people are oblivious of the dangers involved with such often incurable diseases. The majority of Blood Cancers are rare, life-threatening illnesses within limited patient populations; together, they account for 7% of all malignancies. Patients may feel abandoned and have difficulty finding the necessary assistance and information due to the complex, often sparse nature of Leukemia. When it comes to Acute Leukemia, if therapy is not started on time, the patient might succumb to the ailment within a few months. It is vital to diagnose Cancer be it of any type, in its early stages to ensure timely treatment and increase chances of survival. Detecting Leukemia manually in labs by medical personnel examining blood samples is a time and resource-consuming procedure. Customarily, the patients suffering do not have the liberty to exhaust their time as they need immediate care. We need systems that can make use of the latest technological developments in artificial intelligence to produce expeditious and more accurate results.

Keywords: Convolutional Neural Network, Support Vector Machine (SVM), Image Processing, neural networks, decision trees.

Abstract: The current dominating communication system is 4G. However, with the increase in the data rate and in the number of users in the world, the 4G communication system has started to saturate and couldn’t manage to keep up with user demands and there is less room for progress at 4G systems. In search of finding a system that covers the future interests of users, a new communication scheme is being processed as 5G. The objective of this research is to explore, develop and characterize advanced substrates for 5G applications in the frequency range of 26-33 GHz.

Keywords: Diplexer, 5G, symmetric system.

Abstract: Load balancing is one of the major issues in mobile Ad hoc Network (MANET) because of the frequent changes in topology. Load balancing approach is required to remove the possibility of congestion by that the whole traffic is affected in network. The multipath protocol like AOMDV is able to handle the problem of congestion in network because alternative path is available in network to handle load on a particular link or node but only multipath are not gives the satisfactory results. The queue capacity of nodes is limited in network and if full then in that case the packets are drop it means it requires some more queue size. In this research we proposed a new queue estimation approach of load balancing with AOMDV protocol. In this approach the dynamic queue variation scheme is applied to all nodes in network to improve the storing and forwarding capacity of nodes. If the number of nodes are required the some extra queue length then in that case it is available in proposed scheme. The normal AOMDV protocol are not individual suitable for load balancing that is proved by comparing the result of proposed scheme with normal AOMDV. The actual performance of proposed scheme are improves the 13% network performance as compare to AOMDV.

Keywords: MANET, Multipath, Congestion, AOMDV, Dynamic Queue, Load Balancing

Abstract: Relevance feedback is a valuable technique in improving the performance of information retrieval systems.  Many techniques are used nowadays to evaluate keyword relevance in which the utilization of particular keywords are estimated to see if they have been used extremely. Its role is to enhance the search results quality by user involvement in the process and adapting the system based on user feedback. Relevance Feedback provides benefits like relevance improvement, query refinement, reduction of user effort.

Abstract: The integration of Artificial Intelligence (AI) generation into organization operations indicates a huge shift in current-day commerce, reshaping techniques, desire-making techniques, and regular strategic planning methodologies. This paper explores the profound effect of AI on commercial enterprise leadership and strategy, addressing the demanding conditions, opportunities, and satisfactory practices vital for the powerful implementation of this transformative technology. With the upward push of AI, organizations throughout numerous industries are compelled to include their capabilities to stay competitive and progressive. From leveraging good-sized datasets to improving computational power, the adoption of AI is catalyzed by different factors, permitting companies of all sizes to harness its potential for a couple of causes. Industry leaders like Amazon, Netflix, and Google exemplify how AI can revolutionize purchaser opinions, optimize operations, and force innovation. Moreover, AI's transformative energy extends at some point to advertising, operations, finance, and human sources, permitting predictive renovation, custom-designed patron engagement and efficient supply chain manipulation. As AI transforms commercial organization operations, leaders should navigate moral issues and foster collaboration among AI experts and area professionals. By imposing sturdy governance frameworks and ethical hints, agencies can mitigate dangers and make certain accountable AI deployment. Embracing strategic approaches and fostering a lifestyle of innovation allows organizations to leverage AI's transformative capability and pressure sustainable business company achievement within the evolving landscape of AI technology.

Abstract: Large and complex datasets in the quickly changing field of healthcare defy conventional research methods due to their sheer number, minute details, and fast-paced nature. Techniques that can efficiently handle and analyze large datasets—including clinical, personal computer, and medical usage data—are desperately needed. Conventional statistical models face difficulties because, in spite of their vastness, these datasets are either incomplete or restricted to particular segments of the population. Although machine learning approaches have demonstrated their ability to overcome these obstacles, they are not impervious to the biases that are frequently present in observational studies. For these models to be reliable and accurate in research applications, they must be rigorously validated using industry-standard testing techniques like lasso or ridge regression.

Keywords: Machine Learning (ML), Healthcare, Deep Learning, Big Data, Artificial Intelligence (AI), Predictive Analytics, Supervised Learning, Unsupervised Learning, Reinforcement Learning, Data Mining