基于改进人工蜂群的模糊C-均值聚类算法

模糊C-均值聚类算法在数据挖掘领域有着广泛的使用背景,而对初始点的敏感和较差的搜索能力,限制了算法的进一步推广应用。人工蜂群算法具有对初始点不敏感、适应能力强和搜索能力强等优点,并且针对人工蜂群算法对单峰问题收敛速度慢、多峰问题容易陷入局部最优等问题,通过引入差分进化算法中变异和交叉思想,改善蜂群算法的收敛速度,平衡局部搜索和全局搜索能力。然后将改进的人工蜂群算法和模糊C-均值聚类算法结合得到基于改进人工蜂群的模糊C-均值聚类算法,并在多个国际标准数据集上进行验证,实验结果表明此算法在多个衡量指标上取得了明显的改进。

A fuzzy C means clustering algorithm based on improved artificial by colony

The fuzzy C means clustering algorithm has a wide range of applications in data mining. Due to its sensitivity to the initial point and poor search ability, further applications of the algorithm are restricted. The artificial bee colony algorithm is not sensitive to the initial point and has remarkable searching ability and adaptability, however, it suffers slow convergence speed in solving one peak problems, and it is easy to fall into local optimum faults in solving multi peak problems. Aiming at these problems, we introduce the mutation and crossover ideas of the differential evolution algorithm, which can improve the convergence speed of the swarm algorithm and balance its global and local search ability. We combine the improved artificial bee colony algorithm with the fuzzy C means clustering algorithm, and run it on a number of international standard data sets, which verifies the proposed algorithm.Key words: