Partitioning Clustering Based on Support Vector Ranking

Qing Peng, Yan Wang, Ge Ou, Yuan Tian, Lan Huang, Wei Pang

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)


Support Vector Clustering (SVC) has become a significant boundarybasedclustering algorithm. In this paper we propose a novel SVC algorithmnamed “Partitioning Clustering Based on Support Vector Ranking (PC-SVR)”,which is aimed at improving the traditional SVC, which suffers the drawback ofhigh computational cost during the process of cluster partition. PC-SVR is divided into two parts. For the first part, we sort the support vectors (SVs) basedon their geometrical properties in the feature space. Based on this, the secondpart is to partition the samples by utilizing the clustering algorithm of similaritysegmentation based point sorting (CASS-PS) and thus produce the clustering.Theoretically, PC-SVR inherits the advantages of both SVC and CASS-PSwhile avoids the downsides of these two algorithms at the same time. Accordingto the experimental results, PC-SVR demonstrates good performance inclustering, and it outperforms several existing approaches in terms of Rand index,adjust Rand index, and accuracy index.
Original languageEnglish
Title of host publicationADMA 2016: Advanced Data Mining and Applications
EditorsJinyan Li, Xue Li, Shuliang Wang, Jianxin Li, Quan Z. Sheng
Number of pages12
ISBN (Electronic)978-3-319-49586-6
ISBN (Print)978-3-319-49585-9
Publication statusPublished - 2016
Event12th International Conference on Advanced Data Mining and Applications 2016 - Gold Coast, Australia, Gold Coast, Australia
Duration: 12 Dec 201615 Dec 2016

Publication series

NameLecture Notes in Computer Science


Conference12th International Conference on Advanced Data Mining and Applications 2016
Abbreviated titleADMA 2016
CityGold Coast


  • support vector clustering
  • support vector ranking
  • partitioning clustering


Dive into the research topics of 'Partitioning Clustering Based on Support Vector Ranking'. Together they form a unique fingerprint.

Cite this