C.Deisy Chelliah

• PhD
• Professor at Thiagarajar College of Engineering

Detecting plagiarism poses a significant challenge for academic institutions, research centers, and content-centric organizations, especially in cases involving subtle paraphrasing and content manipulation where conventional methods often prove inadequate. Our paper proposes FTLM (Fuzzy TOPSIS Language Modeling), a novel method for detecting plagia...

Several of the major issues affecting food productivity are a pest. The timely and precise detection of plant pests is crucial for avoiding the loss of agricultural productivity. Only by detecting the pest at an early stage can it be controlled. Due to the cyclical nature of agriculture, pest accumulation and variety might vary from season to seaso...

Nowadays, vaccination plays the major role in controlling the death rates in COVID-19. However, certain people don’t have a trust on vaccines like Covaxin and Covishield. In order to make the decision on these vaccines like Covaxin and Covishield, decision support system (DSS) is necessary using machine learning approach. Many researchers find the...

Energy is existing everywhere and always being transferred from one state to another, even with the slightest movement of an object or matter. The human body uses potential energy and transfers it into kinetic energy when walking. Harvesting energy from the human locomotion is clean. The aim of this research is to study the biomechanics of a human...

An EMG-based exoskeleton robot system that assist the rehabilitation process of upper limb has been proposed in this project. The exoskeleton has four degree of freedom (DOF) capable of various motion such as shoulder flexion/extension, abduction/adduction and internal/external rotation, and elbow flexion/extension. This project report describes th...

The education system is a collective intelligence system where a group of persons ranging from the students to the management thinks and work together to achieve institutions" goals. The primary goal of every institution is achieving good end-semester examination results. This is achieved through proper training (cost accounting) given by educators...

The rapid growth of technologies not only formulates life easier but also exposes a lot of security issues. With the advancement of the Internet over years, the number of attacks over the Internet has been increased. Intrusion Detection System (IDS) is one of the supportive layers applicable to information security. IDS provide a salubrious environ...

Pre-diabetes is the forerunner stage of diabetes. Pre-diabetes develops type-2 diabetes slowly without any predominant symptoms. Hence, pre-diabetes has to be predicted apriori to stay healthier. The risk factors for pre-diabetes are abnormal in nature and are found to be present in a few negative test samples (without diabetes) of Pima Indian Diab...

Pre-diabetes is the forerunner stage of diabetes. Pre-diabetes develops type-2 diabetes slowly without any predominant symptoms. Hence, pre-diabetes has to be predicted apriori to stay healthier. The risk factors for pre-diabetes are abnormal in nature and are found to be present in a few negative test samples (without diabetes) of Pima Indian Diab...

Biometrics-based authentication is a most needed activity in a corporate and business world. Genuineness, accuracy and reliability are the most common characteristics of any authentication system. This requires any multimodal unique biometric traits combined with better fusion strategy. This paper have proposed C²-based fusion algorithm for combini...

Multi biometric system can be used in cloud computing to achieve higher data security. Biometric authentication refers to automated methods used to identify a person by the features such as face, iris, vein, finger print, palm print etc. In this paper we proposed a novel C2 code derived using orientation and magnitude information extracted from fin...

Frequent pattern mining using sliding window over data streams is commonly used due to its wide applicability. Determining suitable window size and detection of concept change are the major issues and can be addressed by having flexible window based on amount of changes in frequent patterns. For mining frequent patterns over data streams, vertical...

Diabetes is one of the world’s most common chronic disease. Prediabetes is the pre-phase of diabetes, which slowly lead to type-2 diabetes. Early detection of diabetes prevents hazardous health and saves a life. Pima Indian Diabetes dataset found to have a group of data with prediabetes conditions among the patient who have been classified as “diab...

Outlier refers to a sample which are deviated with the rest of the sample in the data set. These outliers can be categorized as global outliers, contextual outliers, and collective outliers based on the behaviour and the degree of its difference from the normal sample. Methods like supervised, unsupervised and semi-supervised are employed to find t...

Anaphora resolution is an important task to be achieved in many natural language understanding (NLU) applications including machine translation. This paper proposes learning-based system to resolve pronouns in Tamil text built around various classification algorithms. To improve learning accuracy, the system is built in two folds. First is feature...

Health care administration is very poor in rural areas and the scenario is predicted to persist so. Due to the ubiquitous usage, mobile phones are greatly employed in telecardiology systems to bridge this gap. Towards accomplishing Electrocardiogram (ECG) analysis on such resource-constrained devices, we propose mobile-Cardiovascular Abnormality De...

A data stream is an input massive data that arrives at high speed and it is unbounded. The sliding window model is used to extract the recent frequent patterns by adjusting the window size containing only the recent transactions and eliminating the old transactions. Another acute challenge in frequent pattern mining is the presence of null transact...

Learning is a cognitive activity which differs from person to person and hence needs personalization. When learning is performed online, the system needs to understand various traits of learner and deliver Learning Objects (LO) suitable for them to achieve personalization. Many research works have been developing personalization strategies based on...

Remote cardiovascular disease (CVD) diagnosis from ECG plays an important role in health care domain. Data mining, the major step in the process of the extraction of knowledge using descriptive and predictive algorithms that aid in making proactive decisions, has also been used for CVD diagnosis. Recently, diverse techniques have been developed for...

Over the past 25 years, Learning Style Models have brought increasing attention to the fact that students have varying learning styles and students may enjoy learning only if teaching learning methodologies adapt to their learning style. There exists a lot of learning style models that determine the learning styles. However, addressing the needs of...

Most of the mining techniques have only concerned with interesting patterns. However, in the recent years, there is an increasing demand in mining Unexpected Items or Outliers or Rare Items. Several application domains have realized the direct mapping between outliers in data and real world anomalies that are of great interest to an analyst. Outlie...

This paper presents a novel Lossless ECG Compression using Symbol substitution (LECS) deployable on low computational devices (LCD) like mobile phones for effective use in telecardiology. Of the few LCD deployable compression algorithms, even losslessly compressed ECG suffers transmission loss in Global System for Mobile (GSM) networks due to the r...

Personalized E-learning, as an intelligent package of technology enhanced education tends to overrule the traditional practices of static web based E-learning systems. Delivering suitable learning objects according to the learners’ knowledge, preferences and learning styles makes up the personalized E-learning. This paper proposes a novel approach...

Most of the data mining algorithms perform analysis on quantitative data only after performing discretization. Nowadays, there is a great interest in finding the health impacts of climate change. One of the factors that cause changes in the climate is the ozone layer. Adverse levels of ozone may cause several diseases like asthma, chronic disorders...

E-learning has been regarded as the future education and focused by many research works to improve its intelligence and effectiveness as a knowledge-based system. Personalised e-learning, is one of such major issues in the field of knowledge-based e-learning systems. Teaching according to the ability and attitude of the learners is one of the major...

The previous work in web based applications such as mining web content, pattern recognition and similarity measures between the web documents. This paper is about, analyzing web documents in an enhanced way and delve the distillation web document will be the next pace in hypertext mining. The sparse document is a very little data on the web, which...

Most of the mining techniques has only concerned with interesting patterns. However, in the recent years, there is an increasing demand of mining the Unexpected Items or Outliers or Rare Items. Several application domains have realized the direct mapping between outliers in data and real world anomalies that are of great interest to an analyst. Out...

In this paper, we present a fuzzy data mining approach for extracting association rules from quantitative data using search tree technique. Fuzzy association rule is used to solve the high dimensional problem by allowing partial memberships to each different set. It suffers from exponential growth of search space, when the number of patterns and/or...

In this paper we present a GA based fuzzy data mining approach for extracting class association rules from quantitative data with optimal membership function. It is not an easy task to know a priori the most appropriate membership function for mining class association rules. To find the optimal membership function a fuzzy based CHC genetic learning...

A good text classifier is a classifier that efficiently categorizes large sets of text documents in a reasonable time frame and with an acceptable accuracy. Most of the text classification approaches are based on the statistical analysis of a term, either a word or a phrase. Though statistical term analysis shows the importance of the term, it is t...

Association rule mining is a highly popular data mining technique which shows the attributes value conditions that occur frequently together in a given dataset. Even though association rule mining finds its applications widely, it results in more number of irrelevant rules, which lead to the degradation in the prediction accuracy and time consumabl...

Web search engines are essential nowadays as if we, the end users want to know or gather certain information on any field is done by web. It’s a common tendency that people don’t prefer waiting for an information or searching all down many pages to obtain certain information. But as the information growth is tremendous the result set obtained by th...

The inclusion of irrelevant, redundant, and inconsistent features in the data-mining model results in poor predictions and high computational overhead. This paper proposes a novel information theoretic-based interact (IT-IN) algorithm, which concerns the relevance, redundancy, and consistency of the features. The proposed IT-IN algorithm is compare...

Text categorization is a task of automatically assigning documents to a set of predefined categories. Usually it involves a document representation method and term weighting scheme. This paper proposes a new term weighting scheme called Modified Inverse Document Frequency (MIDF) to improve the performance of text categorization. The document repres...

Feature selection is used to eliminate irrelevant and redundant features, which improves prediction accuracy and reduces the computational overhead in classification. This paper presents comparison of 3 methods namely fast correlation based feature selection (FCBF), Multi thread based FCBF feature selection and decision dependent -decision independ...