DocumentCode
2778562
Title
A novel semi-supervised fuzzy c-means clustering method
Author
Li, Kunlun ; Cao, Zheng ; Cao, Liping ; Zhao, Rui
Author_Institution
Coll. of Electron. & Inf. Eng., Hebei Univ., Baoding, China
fYear
2009
fDate
17-19 June 2009
Firstpage
3761
Lastpage
3765
Abstract
In this paper we propose a novel semi-supervised fuzzy c-means algorithm. We introduce a seed set which contains a small amount of labeled data. First, generating an initial partition in the seed set, we use the center of each partition as the cluster center and optimize the objective function of FCM using EM algorithm. Experiments results show that, our method can avoid the defect of fuzzy c-means that is sensitive to the initial centers partly and give much better partition accuracy.
Keywords
expectation-maximisation algorithm; pattern clustering; EM algorithm; FCM algorithm; cluster center; objective function; seed set; semisupervised fuzzy c-means clustering method; Clustering algorithms; Clustering methods; Computer vision; Data analysis; Data mining; Educational institutions; Information retrieval; Mechanical engineering; Medical treatment; Partitioning algorithms; EM; Fuzzy c-means; Semi-supervised;
fLanguage
English
Publisher
ieee
Conference_Titel
Control and Decision Conference, 2009. CCDC '09. Chinese
Conference_Location
Guilin
Print_ISBN
978-1-4244-2722-2
Electronic_ISBN
978-1-4244-2723-9
Type
conf
DOI
10.1109/CCDC.2009.5191706
Filename
5191706
Link To Document