Chaotic gradient artificial bee colony for text clustering

Abstract

Text clustering is widely used to create clusters of the digital documents. Selection of cluster centers plays an important role in the clustering. In this paper, we use artificial bee colony algorithm (ABC) to select appropriate cluster centers for creating clusters of the text documents. The ABC is a population-based nature-inspired algorithm, which simulates intelligent foraging behavior of the real honey bees and has been shown effective in solving many search and optimization problems. However, a major drawback of the algorithm is that it provides a good exploration of the search space at the cost of exploitation. In this paper, we improve search equation of the ABC and embed two local search paradigms namely chaotic local search and gradient search in the basic ABC to improve its exploitation capability. The proposed algorithm is named as chaotic gradient artificial bee colony. The effectiveness of the proposed algorithm is tested on three different benchmark text datasets namely Reuters-21,578, Classic4, and WebKB. The obtained results are compared with the ABC, a recent variant of the ABC namely gbest-guided ABC, a variant of the proposed methodology namely chaotic artificial bee colony, memetic ABC, and conventional clustering algorithm K-means. The empirical evaluation reveals very encouraging results in terms of the quality of solution and convergence speed.