November 2015 Issue Vol.5 No.11
Ph.D (Research scholar), Department of Computer Science, Erode Arts & Science College (Autonomous), Erode, Tamil Nadu, India
Research Supervisor & Head, Department of Computer Science, Erode Arts & Science College (Autonomous), Erode, Tamil Nadu, India
Abstract: Identity authentication is the most essential and required task in the real world environment which make use of biometric model to authenticate the persons in the real time. Biometrics usage to authenticate the system requires is a more burden process where the existences of fake images are exists. In the existing work, this problem is over come by introducing the multi model based authentication system in which wrong authentication due to fake images is reduced considerably. In the existing work, score level fusion is done using the triangulation based method which will combine the multi bio metrics (dual iris, thermal face and normal face) to authentication single person. However the authentication of system based on identity information might be more complex in case of presence of noises present in the given input images. Due to the noises resides in the images, the efficient and accurate matching of test input image with the database images might fail.The performance of combined multi bio metric authentication is improved by replacing the triangulation based method with min-max score level fusion approach which will lead to efficient processing. Thermal image enhancement is done before feature extraction to provide the optimal compilation of finding the fake images. The experimental tests have been conducted in the Matlab simulation environment which provides a flexible and convenient environment for the testers to execute the system. The performance evaluation conducted were proves that the proposed methodology provides better result than the existing system in terms of improved accuracy and successful authentication system.
Keywords: DCT, WALSH, HAAR, RCF.
High Dimensional Feature Based Word Pair Similarity Measuring For Web Database Using Skip-Pattern Clustering Algorithmhttps://archive.org/download//v5no11102/vol5no11102.pdf
Associate Professor, Department of Computer Science, Erode Arts & Science College (Autonomous), Erode, Tamil Nadu, India
M.Phil Scholar (Part Time), Department of Computer Science, Erode Arts & Science College (Autonomous), Erode, Tamil Nadu, India
Abstract: Measuring the semantic similarity between words is an important component in various tasks on the web such as relation extraction, community mining,document clustering, and automatic metadata extraction. Accurately measuring the semantic similarity between words is an important problem in web mining, information retrieval, and natural language processing. In information retrieval, one of the main problems is to retrieve a set of documents that is semantically related to a given user query. Text processing plays an important role in information retrieval, data mining, and web search. In text processing, the bag-of-words model is commonly used. In this paper a new scheme proposes an empirical method to estimate semantic similarity using page counts and text snippets retrieved from a web database for two words.Specifically, it defines various word co-occurrence measures using page counts and integrates those with lexical patterns extracted from text snippets.
Keywords: Web mining, Text processing, Sematic Similiarity, Pattern matching, Skip Lexical Pattern.