期刊刊名:吳鳳學報 卷期:19期
篇名出版日期:2011年12月1日
作者:Chia-Chong Chen
語言:English
關鍵字:Particle swarm optimization, Data clustering, Evolutionary algorithm
被點閱次數:2次
閱讀時間:495sec
摘要: In this paper, a PSO-based clustering method is proposed in an unsupervised manner to automatically determine the appropriate clusters for the considered data set. In the proposed PSO-based clustering method, each individual in the population is constructed to determine a number of clusters and their corresponding cluster centers. Consequently, an individual corresponds to a classification result for the considered data set. Subsequently, a fitness function is defined to guide the searching procedure to select an appropriate number of clusters and their corresponding cluster centers so that the considered data set is efficiently partitioned into the compact and well-isolated clusters. Finally, several examples are considered to illustrate the effectiveness of the proposed PSO-based clustering method.
[ 關閉視窗 ]