An Improved Genetic Clustering Algorithm for Categorical Data

• Main Authors: Jasni, Mohamad Zain, Qin, Hongwu, Ma, Xiuqin, Herawan, Tutut
• Other Authors: Washio, Takashi
• Format: Book Section
• Language: English
• Published: Springer 2013
• Subjects: QA75 Electronic computers. Computer science
• Online Access: http://umpir.ump.edu.my/id/eprint/6186/1/PAKDD13.pdf; http://umpir.ump.edu.my/id/eprint/6186/; http://dx.doi.org/10.1007/978-3-642-36778-6_9
• Institution: Universiti Malaysia Pahang

Deng et al. [Deng, S., He, Z., Xu, X.: G-ANMI: A mutual information based genetic clustering algorithm for categorical data, Knowledge-Based Systems 23, 144–149(2010)] proposed a mutual information based genetic clustering algorithm named G-ANMI for categorical data. While G-ANMI is superior or comparable to existing algorithms for clustering categorical data in terms of clustering accuracy, it is very time-consuming due to the low efficiency of genetic algorithm (GA). In this paper, we propose a new initialization method for G-ANMI to improve its efficiency. Experimental results show that the new method greatly improves the efficiency of G-ANMI as well as produces higher clustering accuracy.