Improved Method for Pattern Discovery in Text Mining

Abstract

The proposed technique uses two processes, pattern deploying and pattern evolving, to refine the discovered patterns in text documents. The experimental results show that the proposed model outperforms not only other pure data mining-based methods and the concept based model, but also term-based state-of-the-art models, such as BM25 and SVM-based models. PBTM firstly generates pattern based topic representations to model user's information interests with multiple topics; then PBTM selects quality patterns for estimating the relevance of documents. The proposed approach incorporates the semantic topics from topic modeling and the specificity of the representative patterns. The proposed model has been evaluated by using RCV1 and TREC topics for the task of information filtering. Comparing with the state-of-the-art models, PBTM demonstrates excellent strength on document modeling and relevance ranking. Keywords: Pattern mining, Text mining, Text classification