Authors: Qingshan Zhao; Guoyan Meng; Zhijian Wu
Addresses: Department of Computer Science and Technology, Xinzhou Teachers University, XinZhou, Shanxi Province, 034000, China. ' Department of Computer Science and Technology, Xinzhou Teachers University, XinZhou, Shanxi Province, 034000, China. ' State Key Laboratory of Software Engineering, Wuhan University, Wuhan, Hubei Province, 430072, China
Abstract: Attribute reduction is the important part in rough set theory. Enlightened by bacterial foraging processing, this paper combines the idea of bacterial foraging algorithm with particle swarm optimisation and proposes a new algorithm-BFPSO algorithm. In this algorithm, the chemotaxis of bacterial foraging can guide the particles to evolve towards much better direction, in turn, the convergence speed and optimisation capabilities are increasing by using PSO. The proposed algorithm is applied to the attribute reduction. Experiments show that attribute reduction based on BFPSO algorithm achieve much better result in optimisation capabilities by comparing with other algorithms, and show that the better minimal attribute reduction can been found by the algorithm.
Keywords: rough sets; bacterial foraging; optimisation algorithms; BFOA; particle swarm optimisation; PSO; attribute reduction; rough set theory; chemotaxis.
International Journal of Innovative Computing and Applications, 2012 Vol.4 No.2, pp.100 - 108
Available online: 07 May 2012 *Full-text access for editors Access for subscribers Purchase this article Comment on this article