Effective pattern discovery for taxt mining using PTM and PDM

摘要

Text mining is the process of discovery of interesting knowledge in text documents, many data mining techniques have been proposed for mining useful patterns in text documents. It is challenging issue to accurate knowledge in [1] text documents to help uses to find what they want. In existing information system they provided many term based methods to solve this challenge. As we know the term-based methods suffers from the problem of polysemy and synonymy. The polysemy means the word has multiple meanings and synonymy is multiple words having the same meaning. In propose to use pattern or phrase based approach so it will perform better than the term based approach. In the proposed approach can improve the accuracy of evaluating term weights because discovered patterns or most specific than whole documents. In this paper we focus on innovative and effective pattern discovery techniques as well as architecture of proposed system. Then we will study how pattern taxonomy model and deploying model is useful for the effective pattern discovery for Text Mining.