A topic-specific data filtering framework based on rough set theory

With the tremendous growth in the volume of text documents available on the Internet and digital libraries, accurate specific topic text filtering is needed. In this paper we propose a rough set aided method to reduce the dimensionality of feature vectors. In order to extract accurate features, we also provide a novel filtering technique called twice-filtering to treat with two different feature sets: "interkeywords" and "intrakeyword". A simple application of E-mail filtering system based on our topic-specific filtering technology shows that with the incorporation of variant weighting methods and more accurate features extracted, our filtering algorithm can speed up the filtering operation with a high precision and recall.