Faculty Publications
Permanent URI for this communityhttps://idr.nitk.ac.in/handle/123456789/18736
Publications by NITK Faculty
Browse
13 results
Search Results
Item Intelligent Interference Minimization Algorithm for Optimal Placement of Sensors using BBO(Springer, 2020) Naik, C.; Shetty D, P.In wireless sensor networks, the performance metric such as energy conservation becomes paramount. One of the fundamental problems of energy drains is due to the interference of sensors during sensing, transmission, and receiving data. The issue of placing sensors on a region of interest to minimize the sensing and communication interference with a connected network is NP-complete. In order to overcome the existing problem, we have proposed a new work for interference minimization technique for optimal placement of sensors by employing biogeography-based optimization scheme. An efficient habitats representation, objective function derivation, migration, and mutation operators are adopted in the scheme. The simulations are performed to obtain the optimal position for sensor placement. Finally, the energy-saving of the network is compared with and without interference aware sensor nodes placement. © 2020, Springer Nature Singapore Pte Ltd.Item DNS Amplification DNS Tunneling Attacks Simulation, Detection and Mitigation Approaches(Institute of Electrical and Electronics Engineers Inc., 2020) Sanjay; Rajendran, B.; Shetty D, P.DNS is a critical infrastructure service of the Internet that translates hostnames to network IP addresses and vice versa. The criticality of DNS can be evidenced by the fact that all most all organizations and enterprises do not block DNS traffic, as it would eventually stop access to the Internet. As a result, attackers have been exploiting the DNS infrastructure and using it as a launchpad for carrying out various attacks e.g. DoS/DDoS, DNS reflection amplification, DNS tunneling, NXDOMAIN attack, and DNS hijacking, etc. During the historic implementation of DNS protocol, its security was not considered which lead to the exploitation of various vulnerabilities in the DNS infrastructure.This paper brings out the technicalities behind DNS amplification and DNS tunneling attacks and presents a number of countermeasures and mitigation techniques to protect against these attacks and the DNS Infrastructure. © 2020 IEEE.Item Indian stock market prediction using deep learning(Institute of Electrical and Electronics Engineers Inc., 2020) Maiti, A.; Shetty D, P.In this paper, we predict the stock prices of five companies listed on India's National Stock Exchange (NSE) using two models- the Long Short Term Memory (LSTM) model and the Generative Adversarial Network (GAN) model with LSTM as the generator and a simple dense neural network as the discriminant. Both models take the online published historical stock-price data as input and produce the prediction of the closing price for the next trading day. To emulate the thought process of a real trader, our implementation applies the technique of rolling segmentation for the partition of training and testing dataset to examine the effect of different interval partitions on the prediction performance. © 2020 IEEE.Item Detection of Pneumonia from Chest X-Ray Images(Institute of Electrical and Electronics Engineers Inc., 2024) Shetty, S.P.; Mamatha, N.; Shetty, M.; Keerthana, S.; Shetty D, P.Pneumonia is a dangerous which is caused by various viral agents. The diagnosis and treatment of pneumonia can be difficult because of the similarities with other lung diseases, which underscores the importance of chest x-rays for an early detection. This work presents techniques of pneumonia detection implementing CNNs, VGG16 and ResNet152V2 architectures, together with the Gradient Descent optimization method. The model is trained and tested on one of Kaggle's dataset which have 5,836 images that are labeled. This system automatically extract features from the chest X-Ray images and uses Gradient Descent optimization to improve its ability to differentiate between the pneumonia patients and healthy cases. For given dataset, the result provides accuracy of 96.56%, 95.34%, 92.9% and 94.23% for RestNet152V2,CNN,VGG16 and Gradient Descent respectively. Therefore this framework will facilitate to the detection of lung disease for experts and doctors as well. © 2024 IEEE.Item Integrating Link Prediction and Comment Analysis for Enhanced Cyberbullying Detection in Online Social Interactions(Institute of Electrical and Electronics Engineers Inc., 2024) Pal, V.B.; Shetty D, P.With the widespread use of online platforms and social media, live chatting has become an integral part of our digital interactions. Individuals are dedicating excessive time to social media applications, facilitated by the development of the internet, technology, and Social Network services like Facebook, Twitter, Instagram, and WhatsApp. While some users utilize these platforms for relaxation and spreading positivity, others misuse them by disseminating hateful content, leading to negative emotional impacts such as anger, stress, depression, and even suicide. Cyberbullying, characterized by the transmission of threatening or insulting messages, exacerbates this issue. Users frequently engage in unproductive debates, resulting in mutual insults and wasted time. Although various enhanced cyberbullying detection techniques have been developed, their effectiveness, especially in live chat scenarios, remains a topic of interest. This paper proposes a novel approach utilizing similarity scores as social network features to predict the probability of links between individuals engaged in live chat, enabling the application of restrictions on hate speech transmission and also optimization in friend recommendation systems. Additionally, it explores the classification of posts from popular social media pages into hateful and non-hateful categories. By analyzing comments and replies to these posts, insights into public behavior and reactions can be gained, facilitating the implementation of warnings or restrictions for page administrators. © 2024 IEEE.Item Quantum Optimizer Using MOEAD for WSN’s(Springer Science and Business Media Deutschland GmbH, 2024) Kanchan, P.; Shetty D, P.; Attea, B.A.Optimization of Wireless sensor networks is done with respect to several parameters like energy efficiency, coverage, etc. A WSN is an inter-related collection of sensors. A WSN can be Homogeneous or Heterogeneous. All nodes in homogeneous WSNs have comparable characteristics like energy of the nodes or radius of sensing, etc. In Heterogeneous WSNs, some of these properties differ. MultiObjective Opimization (MOO) simultaneously optimizes more than one objective. The Multi Objective Evolutionary Algorithm with Decomposition (MOEAD) splits/decomposes a problem into subproblems and all these subproblems are simultaneously optimized. In classical computing, a bit is usually represented by 0 or 1. In Quantum Computing, a bit is 0, 1 or a superposition of 0 and 1. In our research, we use MOEAD with quantum computing to optimize the multiple goals of network lifetime along with coverage for WSNs. These WSNs can be homogeneous or heterogeneous. We contrast our methodology with some of the standard methodologies. Simulations show the upsides of our methodology over different techniques referenced here. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.Item A Hybrid Approach to Predict Ratings for Book Recommendation System Using Machine Learning Techniques(Institute of Electrical and Electronics Engineers Inc., 2024) Roy, T.; Shetty D, P.A recommender system is a tool that suggests products or services to users based on their preferences and past behavior, enhancing user satisfaction and engagement. Accurate rating prediction is crucial as it directly impacts the system's ability to provide relevant and personalized recommendations, thereby improving the overall user experience. In this study, we introduce an innovative approach to recommendation systems by proposing an Weighted Hybrid Model that combines an Adaptive K-Nearest Neighbors (AKNN) algorithm and Singular Value Decomposition (SVD). The AKNN algorithm dynamically adjusts the number of neighbors based on user rating density, providing a tailored neighborhood size for each user. By incorporating a hybrid similarity measure that combines cosine similarity, Pearson correlation, and Variance Mean Difference (VMD), our AKNN algorithm effectively captures the multifaceted nature of user-item relationships. We further enhance our recommendation model by combining AKNN with SVD through optimized weighting, creating a Weighted Hybrid Model. This model balances the contributions of the AKNN and SVD components, leveraging the strengths of both approaches to minimize prediction errors. Our evaluation results demonstrate that the Weighted Hybrid Model outperforms several algorithms, including standalone KNN with Z-Score, Item-wise Variance-Mean based Recommender System (IVMRS), KNN with RJAC DUB, and Pearson Baseline with Weighted KNN. The Weighted Hybrid Model achieved the lowest Root Mean Squared Error (RMSE) of 1.54491 and Mean Absolute Error (MAE) of 1.17839, indicating superior predictive accuracy. © 2024 IEEE.Item Building Voting Systems for a Fairer Future: Exploring Blockchain based E-voting with Ethereum for National Elections(Institute of Electrical and Electronics Engineers Inc., 2024) Prasad, S.V.; Shetty D, P.; Shankar, B.R.The increasing adoption of Blockchain technology, spurred by the success of cryptocurrencies, has gained substantial traction across various sectors. A notable application of Blockchain technology is in electronic voting (e-voting), where decentralized nodes enhance the security and integrity of the voting process. Traditional voting methods suffer from shortcomings such as result delays, susceptibility to tampering, hijacking and destruction of voting machines. Given the scalability challenges of blockchains, a single blockchain network cannot feasibly cover all constituencies in a country. Therefore, a more effective approach is to implement multiple smaller independent blockchain networks, with each constituency having its own network and blockchain. This paper discusses the concept of one network and one blockchain for one constituency, which can be replicated for every other constituencies in the country to scale up. It explores a web3-based e-voting system utilizing private Ethereum blockchain technology, focusing on a network architecture and the design that features a DApp (Decentralized Application) application with a user-friendly interface for voting in polling booths and a governing Smart Contract. Voters can cast their votes using unique identifiers like Aadhaar or UID credentials. The outcomes of this proposed e-voting system demonstrate promising and viable performance for governmental elections. However, it is advisable to conduct trials in local elections or general body elections within institutions to validate its efficacy and reliability before wider adoption in larger democratic elections. © 2024 IEEE.Item An Approach for Integrating Behavioral Analytics and Machine Learning for Enhanced Cybersecurity(Institute of Electrical and Electronics Engineers Inc., 2024) Shivappa, P.K.; Shetty D, P.Data breaches and cyber threats have evolved into increasingly complex and stealthy forms. Conventional anomaly detection based on rules is ineffective in identifying numerous contemporary attacks. Hence, User Behavior Analysis is performed on the network traffic flow data to comprehend, model, and forecast users' actions. Nevertheless, the diversity of the methods makes their understanding exceedingly complex. Therefore, domain experts use machine learning (ML) to accomplish their goals. Thus, this paper aims to suggest an innovative architecture that can detect anomalies in the network traffic flow by analyzing user behavior. The two different sets of data are used for two-class and four-class classification. Both the data are pre-processed for duplicates, missing values, and performing encoding techniques. The correlation analysis is performed to understand the user's behavior before training the ML models. The four different ML algorithms, like Logistic regression LR, KNN, DT, and RF algorithms are applied to the pre-processed datasets. The Random Forest algorithm outperforms by achieving 100% accuracy on two- and four-class classification. The described behavioral modeling approach updates cyber threat detection to match the needs of the modern, ever-changing threat landscape. © 2024 IEEE.Item Enhanced Medicare Fraud Detection Using Graph Convolutional Networks(Institute of Electrical and Electronics Engineers Inc., 2024) Rakesh, M.; Shetty D, P.This paper explores the applications of Graph neural networks (GNN) for enhancing Medicare fraud detection. Graph convolutional network (GCN) is a type of graph neural network. Governments and insurance companies are continuously adapting new technologies to detect and prevent fraud activities and trying to minimize financial losses and improve services because every year they lose billions of dollars due to Medicare fraud. Machine learning algorithms fail to analyze the graph data structure but Graph neural networks are good at analyzing the complex relational data and they directly integrate with the learning process. Machine learning algorithms are facing scalability and generalization across diverse graphs. GNN works on graph data structure, using unique IDs as nodes in a graph, with edges illustrating their relationships. Graph Neural Networks is used to improve the accuracy and efficiency of fraud detection by learning the complex relational information obtained from providers, beneficiaries, and physicians. We created a graph database based on the healthcare provider dataset. In this graph database, two types of heterogeneous nodes are there that are beneficiary and medicare provider nodes. The connection between the beneficiary and medicare providers is a power edge and the connection between providers is a shared-physician edge. We developed a fraud detection model using both machine learning and graph neural networks. Our Graph convolutional Network (GCN) model performed well compared to the basic machine learning (Logistic regression) model. The complex relationships between provider and beneficiary, provider and physician helped to detect medicare fraud using our model. © 2024 IEEE.