A Review: Comparison of Text Documents for feature selection on the use of side information with data miningtechniques

Abstract

Nowadays, the usage of World Wide Web is growing enormously. The corpuses which are available on this world wide web is not structured as it is scattered data , due to thebroad availability of huge amounts of corpuses, there is a need to convert such data into useful information and knowledge. Therefore, there is need of data mining algorithm.However, the relative importance of this side-information or additional information may be difficult to estimate, especiallywhen part of the information is noisy. In such cases, it may be risky to assimilate side-information into the mining process, as it can either enhance the quality of the representation for mining the process, or can add noise to the process. Hence, we need a principled way to accomplish the mining process, so as to maximize the advantages by using this additional information. The study of this paper is immersed in effective clustering, classifying and mining approach with the help of side information or metadata.
Index Terms: Web mining, Data Mining, Parallel comparison, Clustering, Classification, Frequent Pattern, Visualization.