Title :
Fuzzy multisets and fuzzy clustering of documents
Author :
Miyamoto, Sadaaki
Author_Institution :
Inst. of Eng. Mech. & Syst., Tsukuba Univ., Ibaraki, Japan
fDate :
6/23/1905 12:00:00 AM
Abstract :
Aims at developing a method of fuzzy clustering based on fuzzy multisets. Data clustering has been discussed in relation to information retrieval models and fuzzy multisets provide an appropriate model of information retrieval on the WWW. Fuzzy clustering of fuzzy multisets is thus necessary for application to an information retrieval model. A term-document matrix in which entries are multisets are considered. Two dissimilarity measures on fuzzy multisets are proposed. Two methods of fuzzy c-means using these measures are studied in which calculation of cluster centers is focused upon
Keywords :
fuzzy set theory; information resources; information retrieval; pattern clustering; World Wide Web; cluster centers; data clustering; dissimilarity measures; documents clustering; fuzzy c-means; fuzzy clustering; fuzzy multisets; information retrieval models; term-document matrix; Database languages; Fuzzy systems; Information retrieval; Mathematical model; Relational databases; Search engines; Spatial databases; Systems engineering and theory; Web sites; World Wide Web;
Conference_Titel :
Fuzzy Systems, 2001. The 10th IEEE International Conference on
Conference_Location :
Melbourne, Vic.
Print_ISBN :
0-7803-7293-X
DOI :
10.1109/FUZZ.2001.1008956
