基于量子粒子群优化的改进的模糊C-均值聚类算法

模糊C均值算法（FCM）是一种用于聚类的最流行的技术。不过，传统的FCM使用欧氏距离作为数据集的相似准则，从而导致数据集的划分有相等的趋势。而数据集的形状和簇的密度对聚类性能有高度影响。为了解决这个问题，提出基于簇密度的距离调节因子以修正相似性度量。同时，针对模糊C-均值（FCM）聚类算法对初始聚类中心选择敏感，易陷入局部最优的问题，采用量子粒子群优化算法以获取全局最优解。仿真实验证明，改进的聚类算法（QPSO-FCM-CD）具有良好的性能。

Improved Fuzzy C-Means Clustering Based on Quantum-behave Particle Swarm Optimization Algorithm

Fuzzy c-means（FCM） clustering algorithm is one of the most popular techniques used for clustering. However, the conventional FCM uses the Euclidean distance as the similarity criterion of data points, which leads to limitation of equal parti-tion trend for data sets. And the clustering performance is highly affected by the data structure including cluster shape and cluster density. To solve this problem, a distance regulatory factor which is based on cluster density is proposed to correct the similarity measurement. And, Fuzzy c-means（FCM） clustering algorithm has the shortcomings of being sensitive to the initial cluster cen-ters and being trapped by local optima, to resolve two disadvantages, Quantum-behavior Particle Swarm Optimization（QPSO） is used to get the global optimal solution. Data experimental results show that the improved algorithm（QPSO-FCM-CD） has supe-rior performance.