Example-based Chinese text filtering model

Abstract

This paper briefly describes the background of text filtering and proposes an examplebased Chinese text filtering model. It analyzes the structure of the texts as example, extracts the keywords from the texts by means of the text hierarchical analysis presented in this paper, constructs the user profiles which consist of the keywords above, and then filters the new text collections. Consequently, based on the user feedback, it expands the number of example texts, applies the approach of Latent Semantic Indexing to filter texts, and updates the user profiles to improve the efficiency of text filtering systems.