SAIRAMESH L
@stjosephstechnology.ac.in
ASSOCIATE PROFESSOR
RESEARCH, TEACHING, or OTHER INTERESTS
Computer Engineering, Information Systems
Scopus Publications
Scholar Citations
Scholar h-index
Scholar i10-index
Scopus Publications
S. Indira Priyadharsini, G. Raghuraman, and L. SaiRamesh
Springer Science and Business Media LLC
C. Sureshkumar, S. Sabena, and L. Sai Ramesh
Springer Science and Business Media LLC
V. Sathiyavathi, R. Thenmozhi, L. Sai Ramesh, S. Sabena, and K. Selvakumar
IOS Press
A Mobile Ad-hoc Network (MANET) is a self-constructed network consisting of spatially distributed nodes that cooperatively arrange themselves without any centralized manager or fixed based stations. In MANET, the nodes are deployed in a dynamic scenario, so, the nodes may fail as a result of energy depletion, hardware failure, communication link errors, and malicious attacks. Therefore, it is necessary to design energy-efficient fault-tolerant algorithms and protocols for MANETs. Since the development of Mobile Ad-hoc networks was originally motivated by military applications, such as battlefield surveillance and healthcare applications it is required to have a fast recovery mechanism to overcome the fault condition. In this research work, a fault-tolerant Routing mechanism is designed and implemented to address the fault conditions such as node failure, link failure, and critical battery issues called Fault tolerant cluster based AODV with Error Reporting Routing (CAODVERR) protocol, and to improve the stability of the MANET. Also, an Error reporting mechanism has been incorporated with the Ad-hoc On-Demand Distance Vector (AODV) routing protocol.
A. K. Jaithunbi, S. Sabena, and L. Sairamesh
Springer Science and Business Media LLC
Shaahid Ahmed N, K Selvakumar, Dilshad Begum M, and L. SaiRamesh
IEEE
Prediction and classification of data regarding the economic sustainability status of various districts of Southern Indian States / UT viz., Tamil Nadu, Kerala, Karnataka, Undivided Andhra Pradesh and Pondicherry. Prediction of sustainable employability of population using parameters such as Literacy and Higher Education using Machine Learning approach is the basis of this research work. Since Independence the Government of India has periodically introduced several schemes that aim at generating continuous and sustainable employment to the unemployed youth of large segment of underprivileged communities. Machine learning is an approach where computers are trained to make accurate predictions with previous knowledge and experiences. This project uses Machine learning algorithms to suggest future implementation for the employability of a candidate of a particular district in the recruitment process to sphere head the growth and development of the individual and State.Result analysis is a key for any endeavour. The work given in this paperdiscussed about the approaches of machine learning techniques for predict the employability status of every district of the Southern Indian states. The proposed work considered the data from the census of year 2011, Government of India for conducting the studies of the unemployment. Making betterments in schemes for various districts of different states and Union Territory based on their present employability status is necessary for the overall credibility and successful implementation.
S. Kanimozhi and L. Sairamesh
Institution of Engineering and Technology
B Senthilnayaki, N. Saraswathi, V. Sathiyavathi, M. Shanmuga Priya, and L. SaiRamesh
IEEE
The result of a natural hazard is a natural disaster. Earthquakes, landslides, tsunamis, and volcanoes are all natural disasters that cause economic, environmental, and human damage. The need of the hour is to predict such geological calamities. Furthermore, predicting these calamities is a complicated process that is influenced by a variety of physical and environmental factors. These natural risks can also be predicted using machine learning approaches. This proposed approach is primarily concerned with predicting loss owing to the impacted region. The system was created with the help of well-known classifiers like Decision Tree and Chai Square Technique based feature selection. In comparison to previous classifiers, our proposed classification method produces superior results. The afflicted data set is used to categories, and then the classifiers are compared to determine the disaster’s accuracy.
Asif Jamal G A, K. Kalaiselvi, V. Sathiyavathi, N. Saraswathi, and L. Sai Ramesh
IEEE
Agriculture is the backbone of world economy and recent researchers working on improving the crop yield using recent technologies. This article is the one addresses the crop disease prediction using machine learning approaches which detects the disease in crop before it reaches the severity which may result in loss in crop yield. Machine learning and data mining approaches supports the farmers to identify the crop diseases by the features existent in the crop in early stage itself and make them to take necessary steps to reduce the diseases in the crop. This work targets the grass grub insect which creates loss in crop yield. The system uses different existent machine learning approaches and it combines this algorithm to evaluate the efficiency of prediction accuracy. The obtained results shows that the hybrid or ensemble approaches achieve more accuracy than individually used algorithms.
S. A. Sadhana, K. Selvakumar, S. Sabena, and L. SaiRamesh
IEEE
Various social networks have become important part of the people. This social network can be represented as graph. which change with respect to time different nodes and edges are added. In this graph users can be represented as nodes and edges can be thought as relationship between these nodes (User). The goal of link prediction is to finding missing links and prediction future relationships. The task of recommending new relationships(edges) to users(nodes) can be framed as task to Link prediction in a graph. We will use supervised Machine learning approaches with a set of manually extracted features along with node embedding generated using node2vec to increase the performance. Then the trained model is used for performing link Prediction between nodes on the Twitch dataset, which is collected from SNAP. recommendation problem can be mapped to a binary classification problem wherein the two classes are - recommend or do not recommend. We propose solving this by generating link predictions and then using the predicted links to recommend streamers. Thus, we aim to train classification models using machine learning to predict whether a link exists between any given pair of nodes and use this prediction for recommendations. of game. The performance of the model is assessed utilizing prediction performance metric
Razia Sulthana A, Anukriti Jaiswal, Supraja P, and Sairamesh L
IEEE
Customer segmentation has seen major growth in all sectors in the last decade. Several techniques have been devised to analyze customer behavior through loyalty, purchases, recency, frequency and monetary to develop efficient marketing strategies that cater to each client individually. As the availability of products and services increases, so does the competition. With the spiraling of automation accompanied by its cost-effectiveness and ease of availability, all businesses equip themselves with the required workforce and machinery to conduct experiments such as customer segmentation on an industrial scale. In the proposed work, the datasets are manipulated by extracting features from existing attributes. A widespread approach is RFM that calculates the Recency, Frequency and Monetary values for each customer tuple. This paper aims at laying out a new approach at every step of customer segmentation from pre-processing, clustering, validation and suggesting marketing strategies for customer retention. Two datasets- Online Retail II Set and Mall Customer Segmentation are modelled and the results from analysis of both the datasets are presented and compared to reach a generalized opinion.
Sadhana SA, Sabena S, SaiRamesh L, and Kannan A
SAGE Publications
In the field of marketing, many surveys were conducted to analyze the customer satisfaction on products in their online purchases. But the real view of customers about the product is mirrored in the customer’s online reviews (COR) given by them, while they purchase the product online. This paper is the one for analyzing and distinguishing the real view about the customer satisfaction by reviewing their opinions for the product which they buy. As a part of opinion mining, the polarity of the specific word is extracted and classifies the review as positive or negative using Naïve Bayes classifier. And this creates a genuine view about the product from the customer point of view. The real opinion about the customer view on online shopping is going to be distinguished according to the intelligent rules generated based on the hypothesis. Intelligent rules help to classify the reviews by extracting the real opinion of the customer based on the feature they specified for the product which is purchased by the consumer. This kind of feature-based review classification supports the purchase of new users when they approach online shopping. This work also projects the customer view about which feature they really need and also feel good, from their review representation.
N. Fareena, C. Yogesh, K. Selvakumar and L. S. Ramesh
Korean Society for Internet Information (KSII)
L. SaiRamesh, S. Sabena, and K. Selvakumar
Springer Science and Business Media LLC
K. Selvakumar, P. J. A. Alphonse, and L. SaiRamesh
Chapman and Hall/CRC
Y. Reeginal, N. Anu, M. Archana, V. Sathyavathi, and Sai Ramesh L
IOS Press
The channel allocation is the primary concept for enhancing the throughput and channel quality. The target channel allocation will enhance the performance by minimizing the noise rate as the resource utilization leads to an important problem for primary user. The dynamic channel allocation technique is maintained in this paper through the multi-objective optimization technique. The poor resource allocation leads to desperate problem in the wireless network, the proposed technique shows the improved throughput and energy proficiency.
Senthilnayaki B, Mahalakshmi G, Narashiman D, Mahendran E, PremAnandh M, and Sairamesh L
IOS Press
Secure data and property transmission between two parties is a difficult task in the real world. The existing system for e-governance operations is riddled with inefficiencies and dishonesty, resulting in records that are not safeguarded, and people are the ones who face the brunt of it. The proposed system is built upon utilizing blockchain principles to securely transfer E-governance processes from one party to another. It is an electronic record of data that necessitates the use of digital security. All of the data in the blockchain is immutable; it is almost impossible to modify the value of data once it has been placed into the blockchain. The proposed model is based on blockchain technology and implements a secure E-governance system in land registration. It aids in the upliftment of the disadvantaged and underprivileged sections of society by combating illicit land authorization. Because of this, the suggested system outperforms the current system.
Kanimozhi Soundararajan, Anbarasi Soundararajan, and Sai Ramesh L
IOS Press
Anomaly detection is a challenging task in the surveillance system due to the factors like extracting appropriate features, inappropriate differentiation among the normal vs abnormal behaviours, the sparse occurrence of abnormal activities and environmental variations. In the dark environment, detection of human actions is still difficult as more features for recognizing the key point are not visible. Hence the proposed work is focused on overcoming the environmental variations task that too in a less bright environment by using thermal videos. Variations in the actions can be easily identified as it works on the property of infrared radiations. For recognizing actions, the skeleton-based approach is used as it helps with the joint-wise segregation of human parts, resulting in more accuracy. The motion pattern of humans in the thermal video is tracked to classify the level of abnormality.
S. Vatchala, S. Sasidevi, Dhanalakshmi R, and SaiRamesh L
IOS Press
Object detection is one of the most basic and central tasks in computer vision. object detection is a subset of object recognition. Its task is to find all the interested objects in the image, and determine the category and location of the objects. Object detection is widely used and has strong practical value and research prospects. Applications include face detection, pedestrian detection and vehicle detection. In recent years, with the development of convolutional neural network, significant breakthroughs have been made in object detection. This work aim to detect objects in the video frames. It detects household objects and predicts the object where it may be present. Convolutional Neural Networks (CNN) is used to detect objects in the environment. Then Resnet50 is used to classify the images into objects. Then Support vector machine (SVM) is used to train objects and stored in object database. It will be retrieved whenever neural networks sent object for verification.
A. K. Jaithunbi, S. Sabena, and L. SaiRamesh
Springer Science and Business Media LLC
L. Sai Ramesh, S. Shyam Sundar, K. Selvakumar, and S. Sabena
World Scientific Pub Co Pte Lt
Usage of the internet is increasing in the daily life of humans due to the need for speedy task completion for their daily services. Most of the living time is spent in some indoor environment which provides WiFi which is the basic need of internet connectivity using Wireless Access Points (WAP). Nowadays, most of the devices are IoT-based ones, which connect with the outer world through the access points in the existing environment. The wearable IoT devices may be misplaced somewhere and we need a specific scenario which helps to identify the misplaced mobile devices based on access points where they are connected by their unique identity such as MAC address. Most of the time, unrestricted WiFi access provided in the public environment is used by the end-user. In that scenario, the tracking of misplaced mobile devices is creating an issue when the WiFi is in switch-off mode. This paper proposes a technique for tracking a mobile device by using a location-aware approach with KNN and intelligent rules by tracking the channel accessed by the user to find the misplaced path by examining the device connected WAP positions.