论文信息 - An improved artificial bee colony algorithm for clustering

An improved artificial bee colony algorithm for clustering

Artificial Bee Colony (ABC) algorithm, which was initially proposed for numerical function optimization, has been increasingly used for clustering. However, when it is directly applied to clustering, the performance of ABC is lower than expected. This paper proposes an improved ABC algorithm for clustering, denoted as EABC. EABC uses a key initialization method to accommodate the special solution space of clustering. Experimental results show that the evaluation of clustering is significantly improved and the latency of clustering is sharply reduced. Furthermore, EABC outperforms two ABC variants in clustering benchmark data sets.

Hejun Wu | Xingcheng Liu | Biao Hu | Qiuhang Tan

[1] Mingru Zhao,et al. Data Clustering Using Particle Swarm Optimization , 2014 .

[2] Anil K. Jain,et al. Data clustering: a review , 1999, CSUR.

[3] Dervis Karaboga,et al. A powerful and efficient algorithm for numerical function optimization: artificial bee colony (ABC) algorithm , 2007, J. Glob. Optim..

[4] B. Kulkarni,et al. An ant colony approach for clustering , 2004 .

[5] Dervis Karaboga,et al. A novel clustering approach: Artificial Bee Colony (ABC) algorithm , 2011, Appl. Soft Comput..

[6] Andries Petrus Engelbrecht,et al. Data clustering using particle swarm optimization , 2003, The 2003 Congress on Evolutionary Computation, 2003. CEC '03..