Adjustable preference affinity propagation clustering

详细信息    查看全文

• 作者：Ping Li ; Haifeng Ji ; ^{hfji@iipc.zju.edu.cn} ; Baoliang Wang ; Zhiyao Huang ; Haiqing Li
• 关键词：Pattern recognition ; Clustering analysis ; Affinity propagation ; Factor graph
• 刊名：Pattern Recognition Letters
• 出版年：2017
• 出版时间：1 January 2017
• 年：2017
• 卷：85
• 期：Complete
• 页码：72-78
• 全文大小：957 K
• 卷排序：85

文摘

An Adjustable Preference Affinity Propagation (APAP) algorithm is proposed. The initial value of each element preference pk is determined according to the data distribution. Element preference pk can be automatically adjusted during the iteration process. Experiments on synthetic and real data verify the effectiveness and the advantages of APAP.