基于聚类的无监督模型的分类

A comprehensive comparative study of clustering-based unsupervised defect prediction models

这篇论文将40个无监督模型分成了9个聚类族,帮助我(初学者)建立基于聚类的无监督模型的初步印象,然而缺乏直观的认识。在搜索无监督模型的过程中,我发现有一些十分生动的GIF图描述这些模型的聚类过程,因此,我打算将二者组合起来。

 1.基于分区的聚类族/PBC/Partition Based Clustering family)

K-Means (Hartigan and Wong,1979),

 

Cascade K-Means(CM) (Karegowda等人,2012),

 

Canopy (McCallum等人,2000),

 

X-Means(Peleg等人,2000),

 

K-Medoids (Jin and Han,2016),

 

围绕Medoids的划分(Partitioning Around Medoids / PAM)(Kaufman and Rousseuw,2009),

 

Mini Batch K-Means(MBM),(Alonso,2013),

 

Fuzzy C-Means (FCM) (Bezdek et al., 1984),

 

Fuzzy C-Shell (FCS) (Dave, 1990),

 

Hard C-Means (HCM) (MacQueen et al., 1967),

 

K-Modes (Huang, 1997),

 

FarthesFirst (FF) (Hochbaum and Shmoys, 1985),

 

Clustering LARge Applications (CLARA) (Kaufman and Rousseeuw, 2009).

 

上一篇:python——k-means算法


下一篇:GBK均值聚类算法【含Matlab源码】