Hybrid Clustering Based on a Graph Model

Research output: Contribution to book or proceedingConference articlepeer-review

Abstract

A hybrid clustering approach is proposed for processing image-like data such as plots in flow cytometry. Clustering or partitioning data into relatively homogeneous and coherent subpopulations can be an effective pre-processing method to achieve data analysis tasks such as pattern recognition and classification. Our method uses a graph to model the initial manual partition of the dataset. Based on the graph model, an algorithm is developed for automatic detection of regions defined by the partition. A clustering algorithm using Markov Chain Monte Carlo method is developed for finding optimal adjustments to the partition automatically.

Original languageEnglish
Title of host publicationProceedings - 2016 9th International Symposium on Computational Intelligence and Design, ISCID 2016
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages242-245
Number of pages4
ISBN (Electronic)9781509035588
DOIs
StatePublished - Jul 2 2016
Event9th International Symposium on Computational Intelligence and Design, ISCID 2016 - Hangzhou, Zhejiang, China
Duration: Dec 10 2016Dec 11 2016

Publication series

NameProceedings - 2016 9th International Symposium on Computational Intelligence and Design, ISCID 2016
Volume1

Conference

Conference9th International Symposium on Computational Intelligence and Design, ISCID 2016
Country/TerritoryChina
CityHangzhou, Zhejiang
Period12/10/1612/11/16

Scopus Subject Areas

  • Computer Vision and Pattern Recognition
  • Artificial Intelligence
  • Computer Science Applications
  • Control and Optimization
  • Modeling and Simulation

Keywords

  • Markov Chain Monte Carlo method
  • clustering
  • image partition
  • machine learning
  • planar graph

Fingerprint

Dive into the research topics of 'Hybrid Clustering Based on a Graph Model'. Together they form a unique fingerprint.

Cite this