DocumentCode
3360635
Title
Weight Computing in Competitive K-Means Algorithm
Author
Cui, Tingting ; Li, Fangshi
Author_Institution
Coll. of Appl. Sci., Beijing Univ. of Technol., Beijing, China
fYear
2012
fDate
11-13 Jan. 2012
Firstpage
430
Lastpage
435
Abstract
This paper presents Weight Computing in Competitive K-Means Algorithm which is derived from Improved K-means method and subspace clustering. By adding weights to the objective function, the contributions from each feature of each clustering could simultaneously minimize the separations within clusters and maximize the separation between clusters. The experiments described in this paper confirm good performance of the proposed algorithm.
Keywords
pattern clustering; competitive k-means algorithm; improved k-means method; objective function; separation between cluster maximization; separation within cluster minimization; subspace clustering; weight computing; Accuracy; Algorithm design and analysis; Clustering algorithms; Complexity theory; Entropy; Partitioning algorithms; Signal processing algorithms; Clustering algorithm; K-means algorithm;
fLanguage
English
Publisher
ieee
Conference_Titel
Computing, Communications and Applications Conference (ComComAp), 2012
Conference_Location
Hong Kong
Print_ISBN
978-1-4577-1717-8
Type
conf
DOI
10.1109/ComComAp.2012.6154887
Filename
6154887
Link To Document