Document clustering is widely used in Information Retrieval\nhowever, existing clustering techniques suffer from local optima problem in\ndetermining the k number of clusters. Various efforts have been put to\naddress such drawback and this includes the utilization of swarm-based\nalgorithms such as particle swarm optimization and Ant Colony\nOptimization. This study explores the adaptation of another swarm\nalgorithm which is the Firefly Algorithm (FA) in text clustering. We\npresent two variants of FA; Weight- based Firefly Algorithm (WFA) and\nWeight-based Firefly Algorithm II (WFAII). The difference between the\ntwo algorithms is that the W FAII, includes a more restricted condition in\ndetermining members of a cluster. The proposed FA methods are later\nevaluated using the 20 News groups dataset. Experimental results on the\nquality of clustering between the two FA variants are presented and are\nlater compared against the one produced by particle swarm optimization,\nK-means and the hybrid of FA and -K-means. The obtained results\ndemonstrated that the W FAII outperformed the WFA, PSO, K-means and\nFA-K means. This result indicates that a better clustering can be obtained\nonce the exploitation of a search solution is improved.
Loading....