A Study on Web Clustering with respect to XiangShan Science Conference

Wen ZHANG and Xijin TANG
Institute of Systems Science, Academy of Mathematics and Systems Science, Chinese Academy of Sciences

Abstract. This paper has presented two clustering results using two different methods to cluster the same Boolean vectors represented the Web documents of XiangShan Science Conference (XSSC). Then, average co-occurrence and average difference are introduced to evaluate the effectiveness of theses two dif-ferent clustering methods. With these two indicators, the evaluation of experi-mental results from these two clustering methods is presented. Also, an ex-tended research on Web clustering is presented in this paper, that is, the automatic concepts generation. At last, the reliability of the automatic concept generation is discussed in this paper.

Keywords: Web clustering, average co-occurrence, average difference, auto-matic concepts generation, Xiangshan Science Conference