新的混合模糊C-均值聚类算法

基于量子行为的粒子群算法(QPSO)是一种改进的粒子群优化算法.它使用的参数个数少,在解的收敛性和全局搜索能力上优于基本的粒子群算法(PSO).将QPSO算法与模糊C-均值(FCM)算法相结合提出一种新的混合模糊C-均值聚类算法(QPSO-FCM),新算法代替了FCM算法的基于梯度下降的迭代过程,在一定程度上克服了FCM算法易陷入局部极小的缺陷,降低了FCM算法的初值敏感度.实验结果表明,改进后的新算法与FCM算法和PSO与FCM结合算法相比,具有良好的收敛性,聚类效果也有较好的改善.

New hybrid fuzzy C-means clustering algorithm

A new hybrid fuzzy clustering algorithm is proposed,which uses quantum-behaved particle swarm optimization (QPSO) al- gorithm and combines with fuzzy C-means (FCM).The QPSO have the less parameters and higher convergent capability of the global optimizing than particle swarm optimization algorithm (PSO).So the iteration algorithm is replaced by the QPSO based on the gradient descent of FCM,which makes the algorithm have a strong global searching capacity and avoids the local minimum problems of FCM in a way.At the same time,FCM is no longer a large degree dependent on the initialization values.The simulation result proves that compared with FCM the new algorithm not only has the favorable convergence but also the clustering effect is improved obviously.