基于概率保真度的聚类分析算法及应用

摘要

Clustering analysis is one of the main tasks of data mining, and the existing clustering analysis algorithm mostly based on the traditional similarity measure methods. In this paper, based on investigating the similarity measure theory, a new similarity measure method is proposed, which includes two parts:the total similarity and structural similarity. Furthermore, clustering analy-sis algorithm based on probability fidelity is presented, which is proved to be feasible through an example analysis.%聚类分析是数据挖掘的主要任务之一，而现有的聚类分析算法大多基于传统的相似性度量方法。该文在研究相似性度量的理论基础上，提出了一种新的相似性度量方法，该方法包含两个部分：总量相似度和结构相似度，进而提出了基于概率保真度的聚类分析算法，并通过实例分析证明该算法是合理可行的。