Browsing by Author "Radha N"

Now showing 1 - 20 of 49

A NOVEL NETWORK INTRUSION DETECTION SYSTEM FOR PREVENTING FLOODING ATTACKS PACKET DROPPING ATTACKS IN MANETS USING DEEP LEARNING ALGORITHM (Conference Paper)
(Springer Science and Business Media Deutschland GmbH, 2024-04-30) Deepa V; Radha N
The network intrusion detection System (NIDS) is more required for maintaining security in the networks. On the other hand it faces more challenges in wireless network than compared to wired network. A wireless network with several nodes connected to one another via wireless components like transmitters and receivers is termed as Mobile Ad Hoc Network (MANET). The characteristics of MANETs are unique pattern, bandwidth, energy, unstable topology, and security. Because of this, MANETs are susceptible to an extensive collection of extortions and assaults, including denial of service (DoS), flooding, impersonation, black holes, and gray holes attacks. This research concentrates on attacks in MANET. This study introduces a novel Enhanced Generative Adversarial Network with Bidirectional Long Short-Term Memory and Cross-correlated Convolutional Neural Network (EGAN-BiLSTM-CCNN) model in MANET. This model was deployed in cluster heads (CHs) for IDS based on the local information of nodes. First this work uses network simulator (NS2) to simulate the flooding, packet dropping in MANET environment. The parameters derived from the nodes must be capable to accurately depict network behavior and distinguish the typical and anomalous network activity. Every sampling interval time results in the creation of a training dataset that includes every training instance, network activity during the designated interval, and an indication of the types of attacks that occurred during this interval. The EGAN-BiLSTM-CCNN IDS model deployed within each CH for intrusion detection, achieving a balance between security and performance in MANETs. Next, the developed model is utilized in the cluster header to identify malicious nodes, hence preventing MANET attacks and improving network speed.
ANALYZING DATA MINING ALGORITHMS USING DERMATOLOGY DATASET
(Department of Computer Science at Nehru Arts and Science, 2010-02-06) Radha N; Rubya T
Machine Learning plays a major role in several applications. Machine learning algorithms can be used to classify the data with more accuracy. In this paper, Dermatology Dataset is used and model created using Weka and performance is compared among various classifiers
ANALYZING THE CITATION NETWORKS USING COMMUNITY DETECTION APPROACHES : A REVIEW (Conference Paper)
(Springer Science and Business Media Deutschland GmbH, 2025-01-28) Kiruthika R; Radha N
Community detection (CD) in a citation network is necessary for detecting the relationships and patterns between scientific articles which results to discover the knowledge and a better understanding of significant research works. Citation network analysis is a field of research in which it demonstrates how the academic articles and their citations are interrelated, most influenced and illustrating the trends of research. CD is an important element for studying citation networks and it provides a way to understand the relationships between different elements of the network. The primary goal of CD in citation networks is to identify groups of closely related publications that have common topics or subjects. This may lead to a better understanding of the topic. The goal of this study is to find the communities based on their citation patterns of linked sources in citation networks to understand the implicit structure and connections between the academic papers using different CD algorithms.
AUTHOR-CENTRIC PATTERN DETECTION IN SCOPUS CITATION NETWORK VIA COMMUNITY STRUCTURES (Conference Paper)
(Springer Science and Business Media Deutschland GmbH, 2025-01-24) Kiruthika R; Radha N
Analyzing citation networks provides effective utilization of research such as identifying domain-specific research influencers, encouraging the growth of collaboration within various scientific domains: Discover the essential trending topics to promote scholarly development. Performing Citation network analysis and bibliometric analysis on Scopus as a complex interconnected network dataset provides an in-depth knowledge about network analysis, structures, and Community detection based on their citation patterns. Citation network analysis involves studying the relationships, interconnections, and patterns among scientific publications through references. The importance of community detection in the Scopus network is to find hidden intellectual connections, interdisciplinary relationships, and research trends. The research objectives of this work focus on using Python-based methodologies to analyze the citation patterns among authors in the Scopus citation network. The key findings are to analyze the citation patterns of top authors, top 10 solo contributing authors, and authorship patterns from the Scopus network. The main contribution of this study is to understand evolving research trends and the importance of community detection in academic fields. This study employs the Louvain approach, which identifies eight communities from the Scopus citation network. The goal is to discover communities within the academic Scopus data through bibliometric and network analysis, with the aim of enhancing research publication in different fields.
BIOMETRIC TEMPLATE SECURITY USING NONINVERTIBLE AND DISCRIMINABLE CONSTRUCTIONS
(VLB Janakiammal College for Engineering and Technology, 2010-01-23) Radha N; Pranitha R.S
Biometric authentication is a technology that measure and analyzes human physical and behavioral characteristics for recognition and authentication to provide security. The main benefit of the biometric technology is that, it is more safe and comfortable than the traditional systems like password and tokens such as smart cards, magnetic stripe cards, photo ID cards, physical keys and can be lost, stolen, duplicated, or left at home. Cancelable biometrics may be a good approach to address the security and privacy concerns on biometric authentication. The security of cancelable biometrics lies on non-invertibility of the transformed templates. So the transforms should be noninvertible and the original template cannot be recovered. In the proposed work fuzzy vault scheme, Fuzzy Vault with Minutiae Descriptors, Irrevocable Cryptographic Key Generation from Cancelable Fingerprint Templates, steganography, cancelable iris biometrics, Palm print based Cancelable Biometric, mobile fingerprint template protection, secure authentication for fingerprint and face are reviewed and implementation is done to secure Biometric template using noninvertible and discriminable constructions.
BODY JOINT AND TRAJECTORY GUIDED 3-D DEEP CONVOLUTIONAL DESCRIPTORS FOR HUMAN ACTIVITY RECOGNITION
(In association with IBM,AICTE,CSIR with Sri Ramakrishna Engineering College, Coimbatore, 2019-07-11) Srilakshmi N; Radha N
Human Activity Identification (HAI) in videos is one of the trendiest research fields in the computer visualization. Among various HAI techniques, Joints-pooled 3D-Deep convolutional Descriptors (JDD) have achieved effective performance by learning the body joint and capturing the spatiotemporal characteristics concurrently. However, the time consumption for estimating the locale of body joints by using large-scale dataset and computational cost of skeleton estimation algorithm were high. The recognition accuracy using traditional approaches need to be improved by considering both body joints and trajectory points together. Therefore, the key goal of this work is to improve the recognition accuracy using an optical flow integrated with a two-stream bilinear model, namely Joints and Trajectory-pooled 3D-Deep convolutional Descriptors (JTDD). In this model, an optical flow/trajectory point between video frames is also extracted at the body joint positions as input to the proposed JTDD. For this reason, two-streams of Convolutional 3D network (C3D) multiplied with the bilinear product is used for extracting the features, generating the joint descriptors for video sequences and capturing the spatiotemporal features. Then, the whole network is trained end-to-end based on the two-stream bilinear C3D model to obtain the video descriptors. Further, these video descriptors are classified by linear Support Vector Machine (SVM) to recognize human activities. Based on both body joints and trajectory points, action recognition is achieved efficiently. Finally, the recognition accuracy of the JTDD model and JDD model are compared
CANCELABLE MULTIMODAL BIOMETRIC AUTHENTICATION SYSTEM WITH DISTORTION ALGORITHM
(International Journal for Research in Technological Studies, 2015-08) Janani B; Radha N
Biometrics refers to authentication techniques that rely on humans physical and behavioral characteristics that can be automatically checked. Biometric based authentication system provides robust security and ease of use than conventional methods of verification system. Multimodal biometric system is one of the major areas of study identified with large applications in recognition system. Unimodal biometric systems challenge with a wide variety of problems such as noisy data, Intra-class variations, non-universality, and spoof attacks. Some of these limitations can be solved in multimodal biometric system. In proposed work, face and fingerprint biometric traits are used for multimodal biometric authentication system. Biometric traits are transformed using distortion algorithm. After the transformation processes pre-processing of images are done to improve the clear visibility of images. The extractions of minutiae feature from fingerprint are achieved using Crossing Number concept and the face features are extracted using the Local Binary Pattern algorithm. To combine both the face and fingerprint features feature level fusion is used. In order to provide additional security to the proposed work the fuzzy vault is introduced by adding duplicate values and having a secret key to lock and unlock the system. Fuzzy vault and distortion acts as an additional layer of security in multimodal biometric user authentication system.
CANCELABLE MULTIMODAL BIOMETRIC USER AUTHENTICATION SYSTEM WITH FUZZY VAULT
(Srisakthi Institute of Engineering, 2016-01-09) Soruba sree S.R; Radha N
Biometrics refers to authentication techniques that rely on humans physical and behavioral characteristics that can be automatically checked. Biometric based authentication system provides robust security and ease of use than conventional methods of verification system. Multimodal biometric system is one of the major areas of study identified with large applications in recognition system. Unimodal biometric systems challenge with a wide variety of problems such as noisy data, Intra-class variations, non-universality, and spoof attacks. Some of these limitations can be solved in multimodal biometric system. In proposed work, face and fingerprint biometric traits are used for multimodal biometric authentication system. Biometric traits are transformed using distortion algorithm. After the transformation processes pre-processing of images are done to improve the clear visibility of images. The extractions of minutiae feature from fingerprint are achieved using Crossing Number concept and the face features are extracted using the Local Binary Pattern algorithm. To combine both the face and fingerprint features feature level fusion is used. In order to provide additional security to the proposed work the fuzzy vault is introduced by adding duplicate values and having a secret key to lock and unlock the system. Fuzzy vault and distortion acts as an additional layer of security in multimodal biometric user authentication system
CANCELABLE TEMPLATE GENERATION BASED ON IMPROVED QUALITY FINGERPRINT IMAGE FOR PERSON AUTHENTICATION
(International Journal of Computer Sciences & Engineering, 2015-01) Janani B; Radha N
Biometric based authentication system provides robust security and ease of use than conventional methods of verification system. Among various biometrics namely iris, face and gait recognition, Fingerprint recognition has been extensively used by several organizations for recognition and authentication purpose because of its low cost, usability and reliable performance. However, the performance of fingerprint identification techniques is extensively depending on the quality of the input fingerprint images. Due to the context of the image-acquisition process, most of the fingerprint images are found to be low or lack in quality. By concerning privacy issues, these types of low-quality fingerprint templates are easily accessible by intruders, thereby lacks the security. To address this issue, the present work proposes quality enhanced, secured biometric template which simultaneously combines the quality enhancement and cancellable template generation techniques for robust authentication purpose. The fingerprint quality can be improved by means of two-phase enhancement technique, learns the acquired input image by enhancing spatial and frequency domain of image respectively. After that, the cancellable fingerprint templates are generated by means of transforming the quality enhanced fingerprint minutiae distortedly by using Distortion Transformation. Experimental results show that the proposed algorithm efficiently holds various input image contexts and attains improved results in terms of quality and security when compared with some state-of-the-art methods, and thus improves the fingerprint-authentication systems performance.
CLASSIFICATION OF USER OPINIONS FROM TWEETS USING MACHINE LEARNING TECHNIQUES
(International Journal of Advanced Research in Computer Science and Software Engineering (IJARCSSE), 2013-07) Poongodi S; Radha N
Online Social Network is a standard platform for collaboration, communication where people are connected to each other for sharing their opinion. In general, opinions can be articulated about anything like products, surveys, topics, individuals, organizations and events. There are two main types of textual information in web like facts and opinions. Facts can be expressed in defined terms by the user implicitly. To mine opinion, from the user defined facts is intellectually very demanding. User opinion is valuable data, which can be used for marketing research in business during decision making process. So opinion mining and classification plays a vital role in predicting what people think about products. In this work, basic Natural Language Processing (NLP) techniques and hash tag segments, emoticons are used for classification. The performance comparison of Support Vector Machine (SVM), Naïve bayes (NB) and Multilayer Perceptron (MLP) are done using weka. It is observed that the MLP gives better accuracy to classify the opinion from tweets
COMPARATIVE STUDY AND ANALYSIS OF AUDIT DATA USING DATAMINING TOOLS
(RVS College of Engineering and Technology, Sulur, 2010-06-10) Radha N; Deepalakshmi T
machine learning for spam classification is an important research issue at present. Classification of Audit dataset is implemented by using Weka, Tanagra, and Rapid Miner. This classification creates a global model, which is used for predicting the class label of unknown data. Thus, it is useful in many decision problems, where for a given data item a decision is to be made. This paper explores and identifies the use of different Machine learning algorithms for classifying the missing and erroneous data in the dataset. Different classification algorithms are experimented and performance is analysed
DIAGNOSIS OF CHRONIC KIDNEY DISEASE USING MACHINE LEARNING ALGORITHMS
(International Journal of Innovative Research in Computer and Communication Engineering, 2016-01) Radha N; Ramya S
Chronic Kidney Disease (CKD) is a gradual decrease in renal function over a period of several months or years. Diabetes and high blood pressure are the most common causes of chronic kidney disease. The main objective of this work is to determine the kidney function failure by applying the classification algorithm on the test result obtained from the patient medical report. The aim of this work is to reduce the diagnosis time and to improve the diagnosis accuracy using classification algorithms. The proposed work deals with classification of different stages in chronic kidney disease according to its severity. The experiment is performed on different algorithms like Backpropagation Neural Network, Radial Basis Function and Random Forest. The experimental results show that the Radial basis function algorithm gives better result than the other classification algorithms and produces 85.3% accuracy
EFFECT OF HALL CURRENT ON MHD OSCILLATORY SLIP FLOW WITH VARYING TEMPERATURE AND CONCENTRATION
(Asian Journal of Applied Sciences, 2017) Sumathi K; Arunachalam T; Radha N
An investigation of the effect of Hall current on an unsteady MHD mixed convective oscillatory flowof an electrically conducting fluid through a planar channel filled with saturated porous medium is carried out in this paper. The effect of buoyancy, heat source, thermal radiation, chemical reaction and Hall current are taken into account with slip velocity, varying temperature and concentration at the lower boundary. A series solution is found using perturbation techniques. The effects of various parameters on the main and cross flow velocity, temperature, Skin friction, rate of heat and mass transfer are discussed numerically.
EFFECT OF MAGNETIC FIELD ON THREE DIMENSIONAL COUETTE SLIP FLOW PAST POROUS PLATES
(Applied Mathematical Sciences, 2014) Sumathi K; Arunachalam T; Radha N
An analysis of the three dimensional flow of an viscous incompressible fluid between two horizontal porous flat plates separated by a finite distance in a slip flow regime is carried out under following conditions: the fluid is electrically conducting, the free stream velocity is uniform, the plate is subjected to a sinusoidal transverse suction velocity distribution and a magnetic field of uniform strength is applied in the direction normal to the plate. The influences of the various parameters on the main flow and cross flow velocity and skin friction are discussed with the help of graphs
EFFECT OF MAGNETIC FIELD ON THREE DIMENSIONAL FLUCTUATING COUETTE SLIP FLOW PAST POROUS PLATES
(2014) Sumathi K; Arunachalam T; Radha N
An analysis of the three dimensional flow of an viscous incompressible fluid between two horizontal porous flat plates separated by a finite distance in a slip flow regime is carried out under following conditions: the fluid is electrically conducting, the free stream velocity is uniform, the plate is subjected to a sinusoidal transverse suction velocity distribution and a magnetic field of uniform strength is applied in the direction normal to the plate. The influences of the various parameters on the main flow and cross flow velocity and skin friction are discussed with the help of graphs.
AN EFFICIENT SECURE BIOMETRIC SYSTEM USING NONINVERTIBLE GABOR TRANSFORM
(IJCSI International Journal of Computer Science, 2011-09) Radha N; Karthikeyan S
Biometric scheme are being widely employed because their security merits over the earlier authentication system based on records that can be easily lost, guessed or forged. High scale employments and the related template storage have increased the requirement to guard the biometric data stored in the system. Theft of biometric information is a negotiation of the user’s privacy. In Addition, the stolen biometric information can be used to access other biometric systems that have the similar feature provided for the user. Several alternative functions have been identified in literature for creating revocable or noninvertible biometric templates. Although, their security examination either disregards the distribution of biometric features or uses inefficient feature matching. This generally shows the way to unrealistic approximation of security. In this paper a novel approach for the Non-Invertible biometric system is proposed to secure the biometric template. Security of a feature transformation method can be assessed according to the two main factors: i) non-invertibility, and ii) diversity. Non-invertibility represents the complexity in obtaining the original biometric when the secure template is provided and diversity represents the complexity in guessing one secure template when a different secure template created from the identical biometric is provided. The proposed Non-invertible Gabor transform possess both the non-invertible and diversity features which enhances the security of the system to a large extent. The proposed approach is very much resistant to minor translation error and rotation distortion. The experimental result shows the better performance of the proposed technique compared to the existing system
FUSION OF IRIS AND RETINA USING RANK-LEVEL FUSION APPROACH
(International Journal of Computer Technology and Application, 2011) Kavitha A; Radha N
Personal identification and authentication is difficulty in all the systems. Shared secrets like Personal Identification Numbers or Passwords and key devices such as Smart Cards are not presently sufficient in few situations. These traditional tokens-based systems may be easily stolen or lost. Biometrics is the only way of improving the capability to recognize the persons according to the physiological or behavioral features. In many real-world applications, unimodal biometric system suffers from some limitations of noise in sensed data, intra-class variation, inter-class similarities, non-universality and spoof attacks. Multibiometric systems seek to alleviate some of these problems by consolidating the evidence obtained from different sources. These systems help to achieve an increase in performance. This paper focused on developing a multimodal biometrics system, which uses biometrics such as iris and retina. Fusion of biometrics is performed by means of rank level fusion. The ranks of individual matchers are integrated using the borda count, and logistic regression approaches. The developed multimodal biometric system utilizes and Fisher’s Linear Discriminant (FLD) and Principal Component Analysis (PCA) methods for individual matchers (Iris and Retina) identification. The features from the biometrics are obtained by using the Fisher face. The experimental result shows the performance of the proposed multimodal biometrics system.
AN IMPROVED TRUST BASED APPROACH FOR DETECTING MALICIOUS NODES IN MANET
(International Journal of Computer Trends and Technology, 2016-11) Nachammai M; Radha N
Mobile ad hoc network (MANET) is an infrastructure-less network of mobile devices connected wirelessly. MANET is used widely today because of its nature as self-configuring, easy to move independently in any directions. MANET acts like a router and therefore changes its links to other devices frequently. Due to its nature MANET has been used in various applications like Military applications, Wireless Sensor Network and so on. As its infrastructure-less and dynamic nature, it is highly affected by various attacks like black hole attack, gray hole attack, DoS attack and many collaborative attacks. Hence security is the main challenge in MANET. Many existing works has done on the basis of detecting attacks by using various approaches like Intrusion Detection, Bait detection, Cooperative malicious detection and so on. But this Trust based approach mainly focuses on detecting the malicious nodes on the trusted path than the shortest path as discovered by using DSR mechanism
LEAF IDENTIFICATION USING MACHINE LEARNING ALGORITHMS
(In association with IBM,AICTE,CSIR with Sri Ramakrishna Engineering College, Coimbatore, 2019-07-11) Radha N; Rehana Banu H
Plants play an important role to equalize the carbon-oxygen cycle in the earth. Without knowing the importance of valuable plants, the plants are at the extinction. To help the naïve user to know about plants and it is important there is demand to develop a system to classify the plants based on the leaves. Due to the boon of ICT and machine learning algorithms, the leaves can be easily classified. Plant Leaf images are collected in Coimbatore. The main aim of this paper is to classify the leaves using Support Vector Machine (SVM) using KBF kernel, K-Nearest Neighbor, AdaBoost classifiers and also the accuracy obtained in these classifiers are compared. The performance of the models is evaluated using 10-fold cross validation method and the results are discussed. The classifier using SVM and KNN outperforms well than Adaboost classifiers
LEVY FLIGHT-BASED DOVE SWARM OPTIMIZATION AND DEEP NEURAL NETWORK FOR FAKE NEWS DETECTION IN SOCIAL MEDIA (Conference Paper)
(Springer Science and Business Media Deutschland GmbH, 2024-05-23) Padmavathy L.; Radha N; Nithya S
By disseminating fake information, fake news plays a significant part in influencing people’s knowledge and perceptions as well as their decision-making. By incorporating real news into fake news, online forums and social media have encouraged its dissemination. Fake news has become the primary obstacle to having a greater impact in the information-driven environment for determined fakers. Due to a number of features in the dataset, testing on a single dataset in the current system may produce false results. The performance of classifying fake news is reduced. This study introduces the Levy flight-based dove swarm optimization (LDSO) and deep neural network (DNN) algorithms for the detection of fake news. The preprocessing, news-user engagement matrix design, feature selection, and fake news classification method are the primary phases of this effort. Preprocessing is carried out using stemming, stop word removal, and tokenization on the features from BuzzFeed and PolitiFact datasets. It is used to eliminate extraneous features to increase the accuracy of predicting fake news. The news-user engagement matrix is then constructed for detection. Latent representation of together the news content and the social context is obtained with tensor by a coupled matrix-tensor factorization algorithm. These features are included in the process of feature selection, which is carried out by the LDSO algorithm. Then DNN method is used to classify fake news by dealing with a variety of filters across every dense layer by dropout. Based on the experimental findings, it was determined that the proposed LDSO-DNN algorithm outperforms the existing methods by increased precision, recall, F-measure, and accuracy.