数据挖掘中聚类算法研究进展_周涛

代码说明：

  聚类分析是数据挖掘中重要的研究内容之一，对聚类准则进行了总结，对五类传统的聚类算法的研究 现状和进展进行了较为全面的总结，就一些新的聚类算法进行了梳理，根据样本归属关系、样本数据预处理、 样本的相似性度量、样本的更新策略、样本的高维性和与其他学科的融合等六个方面对聚类中近 20多个新算 法，如粒度聚类、不确定聚类、量子聚类、核聚类、谱聚类、聚类集成、概念聚类、球壳聚类、仿射聚类、数据流聚 类等，分别进行了详细的概括。(Clustering analysis is one of the important research contents in data mining. The clustering criteria are summarized. The research status and progress of five traditional clustering algorithms are summarized comprehensively. Some new clustering algorithms are used. By combing, according to the sample attribution relationship, sample data preprocessing, sample similarity measure, sample update strategy, sample high-dimensionality and fusion with other disciplines, nearly 20 new algorithms in clustering, Such as granular clustering, uncertain clustering, quantum clustering, nuclear clustering, spectral clustering, clustering integration, concept clustering, spherical shell clustering, affine clustering, data stream clustering, etc., respectively Generalization.)

文件列表：

数据挖掘中聚类算法研究进展_周涛.pdf, 1957907 , 2019-05-11