Asian Science Citation Index is committed to provide an authoritative, trusted and significant information by the coverage of the most important and influential journals to meet the needs of the global scientific community.  
ASCI Database
308-Lasani Town,
Sargodha Road,
Faisalabad, Pakistan
Fax: +92-41-8815544
Contact Via Web
Suggest a Journal
Articles by T. Ravi
Total Records ( 4 ) for T. Ravi
  A. Natarajan and T. Ravi
  Gene selection is very important in classification of cancer using parallel computing in the analysis of gene expression relationship. The high performance parallel computing is used for gene expression analysis and finding the thousands of genes simultaneously. DNA microarrays are used to measure the expression levels of thousands of genes simultaneously. The classification and validation of molecular biomarkers for cancer diagnosis is an important problem in cancer genomics. The microarray data analysis is very much important to extract biologically useful data from the huge amount of expression data to know the current state of the cell. Most cellular processes are regulated by changes in gene expression. This is a great challenge for computational biologists who see in this new technology the opportunity to discover interactions between genes. In this study, we propose a Cooperative Parallel Multi-Objective Genetic algorithm for Gene Feature Selection. We have implemented CPMGA for gene feature selection to classify the breast cancer data sets. More importantly, the method can exhibit the inherent classification difficulty with respect to different gene expression datasets, indicating the inherent biology of specific cancers.
  G. Bharathi Mohan and T. Ravi
  With the growing trend of e-commerce sites, blogs and web forums, people are keenly articulating their opinion on various products, topics. If we are buying a product for the first time, we would go through reviews which are already presented by the users who have used it. Manual analysis can be difficult and consumes more time, thus, a method is required to present the summary of the reviews. Reviews recorded by the users are unstructured in nature. Opinion mining is a discipline of web content mining which in turn is a category of web mining. The other categories of web mining are web structure and web usage mining. Opinion mining can be exploited by both companies and individuals. It involves natural language processing, text analysis and computational linguistics. The focus of the proposed system is mainly in extracting the aspects or features of the product which is the first step of opinion mining. An extension to the Intrinsic and Extrinsic Domain relevance method is made in order to support the rare features too. If the extraction step is improvised, the consequent steps will give fine grained outcomes and thus the result will be enhanced greatly.
  J. Pradeep Kumar , A. Udaya Kumar and T. Ravi
  As the contemporary applications are database-driven, SQL Injection Attacks (SQLIAs) have been capable of causing potential risk to businesses across the globe. Most of the existing solutions focused on SQL and its structure at application level which is doomed to fail when stored procedures are targeted. In this study, we propose a framework for detecting SQLIAs at database level. We exploit kernel level functions and data mining techniques such as classification to have basis for detection of such attacks. The framework provides placeholders to have flexible mechanisms that help in using different approaches in future. Thus, the framework provides pluggable mechanisms, so as to support future techniques as well at database level. We implemented the functionality of the framework using PostgreSQL. The kernel functions of the RDBMS are exploited in order to have integrated functionality to detect SQLIAs. The empirical results revealed that the proposed framework is able to provide 99% probability of protecting applications from SQLIAs. The framework also achieve 100% true positives in detecting SQLIAs.
  J. Pradeep Kumar , A. Udaya Kumar and T. Ravi
  When data is published in the real world it is essential to ensure that privacy is not disclosed and the data is not misused. In our study earlier we proposed an extended misusability measure that helps in finding the probability of misuse of given dataset. The measure takes single or multiple publications as input and generates misusability score. This score determines the level of misusability possible with the given dataset. The misusability leads to possible disclosure of privacy. By computing misusability score, it is possible to anonymize sensitive attributes to achieve privacy preserving data publications and data mining as well. In this study, our aim is to demonstrate the real utility of our extended misusability measure. We proposed a framework with an underlying algorithm to sanitize data before publishing it or before it is subjected to mining. The proposed algorithm employs the measure and determines the need for sanitizing datasets. The algorithm in turn uses K-anonymity which one of the standard sanitization algorithms for preventing privacy attacks on the datasets. We built a prototype application that demonstrates the proof of concept. The empirical results revealed that our misusability measure has significant impact on the privacy preserving data publishing and privacy preserving data mining.
Copyright   |   Desclaimer   |    Privacy Policy   |   Browsers   |   Accessibility