带障碍的量子粒子群聚类算法

Clustering with Obstacles Based on Quantum-behaved Particle Swarm Optimization

典型的基于空间约束的划分聚类算法采用基于梯度下降的搜索方法,存在着易陷入局部极值和对初始值敏感的问题,因此提出带障碍的量子粒子群聚类算法。新算法重新定义了数据点绕过障碍物的距离函数,提出了粒子逃逸原则以避免聚类中心点陷入障碍物中,并且在很大程度上克服了划分聚类的缺点。实验结果证明了该算法的有效性和准确性。

T raditional clustering algorithm based on Space Constraint applies the searching method on gradient descent ,thus it is apt to fall into local extremum and be sensitive to initial parameters .There-fore ,a new clustering with obstructed distance algorithm based on quantum -behaved particle swarm optimization is proposed .The algorithm re-defines the distance function of data points by passing ob-stacles ,applies the Escaping Principle to avoid the updated cluster center particles sinking into the area of the obstacles ,and overcomes the problems of Clustering algorithm .The simulation experiments also illustrate the effectiveness and accuracy of this method .