抽象的

Improved Correlation Preserved Indexing For Text Mining

Vinnarasi Tharania. I,M.Kanchana, V.Kavitha

Data mining is an excellent domain to work where new concepts are implemented. The knowledge extraction and knowledge discovery is a major task of engineering organization. Therefore a new field of study, Knowledge Discovery in Database (KDD), and data mining is explored. And many applications are been developed. In this paper a discussion about field of data mining is also placed. Document clustering is a field where we are implementing a concept of grouping of similar objects. Some groups are formed and the documents are placed under those groups. The main motive of this paper is comparison of various algorithms and giving the best result of the clustering. A new algorithm has been proposed in this paper is ICPI (Improved correlation preserving indexing) which is performed by the correlation similarity measure space. ICPI can successfully find out the essential structures rooted in high dimensional document space. The proposed work is to provide an efficient text mining algorithm to perform mining in the document.ICPI can successfully find out essential structures rooted in high dimensional document space.