Fuzzy clustering validity for spatial data
Tóm tắt
The validity measurement of fuzzy clustering is a key problem. If clustering is formed, it needs a kind of machine to verify its validity. To make mining more accountable, comprehensible and with a usable spatial pattern, it is necessary to first detect whether the data set has a clustered structure or not before clustering. This paper discusses a detection method for clustered patterns and a fuzzy clustering algorithm, and studies the validity function of the result produced by fuzzy clustering based on two aspects, which reflect the uncertainty of classification during fuzzy partition and spatial location features of spatial data, and proposes a new validity function of fuzzy clustering for spatial data. The experimental result indicates that the new validity function can accurately measure the validity of the results of fuzzy clustering. Especially, for the result of fuzzy clustering of spatial data, it is robust and its classification result is better when compared to other indices.
Tài liệu tham khảo
Bezdek J C (1980) A convergence theorem for the fuzzy ISODATA clustering algorithm[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1(2): 1–8
Bezdek J C (1981) Pattern recognition with fuzzy objective function algorithms[M]. New York: Plenum Press
Xie X L, Beni G (1991) A validity measure for fuzzy clustering[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 13(8): 841–847
Fukuyama Y, Sugeno M (1989) A new method of choosing the number of clusters for the fuzzy c-means method[C]. Proceedings of the Fifth Systems Symposium, Japanese
Kim D-W, Lee K H, Lee D (2003) Fuzzy cluster validation index based on inter-cluster proximity[J]. Pattern Recognition Letters, 24(1515): 2 561–2 574
Bezdeck J C, Ehrlich R, Full W (1984) FCM:Fuzzy c-means algorithm[J]. Computers and Geoscience, 23:16–20
Dave R N (1996) Validating fuzzy partitions obtained through c-shells clustering[J]. Pattern Recognition Letters, 17(6): 613–623
Vazirgiannis M, Halkidi M, Gunopulos D (2003) Uncertainty handling and quality assessment in data mining[M]. London, Hong Kong: Springer-Verlag
Pal N R, Bezdek J C (1995) On cluster validity for the fuzzy c-means model[J]. IEEE Transactions on Fuzzy Systems, 3(3): 370–379
Great Basin Center(2007) Nevada-Utah mines and prospects[OL]. http://www.unr.edu/Geothermal/GIS_download3.htm#RRvalNevada_Faults