新的混合小生境鱼群聚类算法

针对K-Means算法对于初始k值较敏感和容易过早收敛的问题，提出基于人工鱼群机制的K-Means聚类算法(NAFS)。首先，利用先验知识随机产生待求解问题的若干个聚类中心，组成一个鱼群环境；其次，利用鱼群个体的协作、竞争机制寻找满意的结果。鉴于人工鱼群算法后期容易陷入局部最优，根据鱼群聚集度引入小生境算法，改善种群的多样性，提高了算法的求解精度。在KDDCUP99数据集上的实验结果表明，该算法具有较高的聚类精度，适用于高维数据的聚类问题。

New clustering algorithm based on hybrid niching artificial fish swarm

To overcome the shortcomings such as being sensitive to initial value of k,premature convergence in K-Means algorithm,this paper presented an improved K-Means algorithm based on artificial fish swarm mechanism named NAFS.Firstly,priori knowledge was exploited to randomly generate some cluster centers for the problems to be solved,then composing the fish swarm environment.Secondly,the cooperation and competition mechanism of fish individuals was utilized to search satisfied outcome.In view of the deficiency that artificial fish swarm is prone to fall in local optimum,niching algorithm was introduced according to the fish crowding density to ameliorate the diversity of population and improve its solution accuracy.The results of experiments on KDDCUP99 show NAFS has higher clustering accuracy and is appropriate to solve clustering problems with high dimensionality.