Title :
Class Assignment Algorithms for Performance Measure of Clustering Algorithms
Author :
Jie Zhang ; Xingsi Xue ; Yuping Wang
Author_Institution :
Sch. of Comput. Sci. & Technol., Xidian Univ., Xi´an, China
Abstract :
To measure the performance or validity of clustering algorithms, several evaluation values, such as successful rate, successful number and full successful rate are defined. In order to ensure each cluster to at least contain one vector data, and to maximize several proposed evaluation values, two class assignment algorithms are designed. To testify their performance, we employ them to the k-means clustering algorithms.
Keywords :
data mining; pattern clustering; class assignment algorithms; data mining; k-means clustering algorithms; performance measure; successful number; successful rate; Algorithm design and analysis; Classification algorithms; Clustering algorithms; Indexes; Partitioning algorithms; Tin; Vectors; class assignment; clustering algorithms; performance measure; successful number; successful rate;
Conference_Titel :
Computational Intelligence and Security (CIS), 2012 Eighth International Conference on
Conference_Location :
Guangzhou
Print_ISBN :
978-1-4673-4725-9
DOI :
10.1109/CIS.2012.31
