期刊名称：International Journal of Computer Information Systems and Industrial Management Applications
印刷版ISSN：2150-7988
电子版ISSN：2150-7988
出版年度：2013
卷号：5
页码：472-479
出版社：Machine Intelligence Research Labs (MIR Labs)
摘要：Good cluster descriptors facilitate the efficient stor- age and retrieval of information. In particular, when the im- precision and uncertainty of textual information is considered, the extraction of cluster descriptors, which represent the com- patibility of a document with a cluster in a more precise way, is a challenging problem. Therefore, in this paper we present the method named Fuzzy-DDE (Fuzzy method for document descriptors extraction), by which two issues are addressed: (1) how to consider the imprecision and uncertainty present in the document clustering and (2) how to extract cluster descriptor from this kind of information. The experimental evaluation shows that the insertion of the information about the compat- ibility of a document with a cluster improves the fuzzy clus- ter descriptor extraction. Furthermore, the proposed method demonstrate its usefulness and effectiveness not only as a de- scriptor extraction, but also as a feature selection method for document categorization.
关键词：fuzzy clustering; text mining; cluster descriptor; infor- ; mation retrieval
