Title :
Evolutionary Composite Attribute Clustering
Author :
Hong, Tzung-Pei ; Song, Wei-Ping ; Chiu, Chu-Tien
Author_Institution :
Dept. of Comput. Sci. & Inf. Eng., Nat. Univ. of Kaohsiung, Kaohsiung, Taiwan
Abstract :
In this paper, we propose a GA-based clustering method for composite-attribute clustering and feature selection. We have also designed a new chromosome representation, a fitness evaluation function, and an adjustment process in the proposed approach. Experimental results show that the proposed approach with composite attributes performs better than that without composite attributes.
Keywords :
data mining; genetic algorithms; pattern clustering; GA-based clustering method; adjustment process; chromosome representation; evolutionary composite attribute clustering; feature selection; fitness evaluation function; Accuracy; Biological cells; Computer science; Educational institutions; Genetic algorithms; Genetics; Machine learning; Classification; Composite Attribute; Feature Clustering; Feature Selection; Genetic Algorithm;
Conference_Titel :
Technologies and Applications of Artificial Intelligence (TAAI), 2011 International Conference on
Conference_Location :
Chung-Li
Print_ISBN :
978-1-4577-2174-8
DOI :
10.1109/TAAI.2011.59
