可能性熵聚类

Possibilistic Entropy Clustering

研究了基于可能性熵理论的聚类问题.首先定义并讨论了可能性熵,继而将可能性熵引入聚类分析,提出了可能性熵聚类算法.它考虑到熵聚类的全局和局部效应,具有清晰的物理意义和数学特征.该算法还能在聚类过程中自动地确定分辨率参数,克服了对于噪声和外围点的敏感性.仿真实验证明,即使各类大小不一,数据集被强噪声所污染时,该算法仍能有效地估计各类中心.

It deals with clustering analysis within the framework of possibilistic entropy theory. First, the possibilistic entropy is defined with brief discussion. Then the Possibilistic Entropy Clustering (PEC) algorithm is developed, which takes into account both global effect and local effect of entropy based clustering and is of clear physical meaning and well-defined mathematical features. Besides, it can automatically control the resolution parameter during the clustering proceeds and overcome the sensitivity to noise and outliers. Simulation experiments show that even when the clusters vary significantly in size and shape, and the data set is contaminated by heavy noise, this novel algorithm can provides efficient and accurate estimation of the cluster centers.