Overview on techniques in cluster analysis

Methods Mol Biol. 2010;593:81-107. doi: 10.1007/978-1-60327-194-3_5.

Abstract

Clustering is the unsupervised, semisupervised, and supervised classification of patterns into groups. The clustering problem has been addressed in many contexts and disciplines. Cluster analysis encompasses different methods and algorithms for grouping objects of similar kinds into respective categories. In this chapter, we describe a number of methods and algorithms for cluster analysis in a stepwise framework. The steps of a typical clustering analysis process include sequentially pattern representation, the choice of the similarity measure, the choice of the clustering algorithm, the assessment of the output, and the representation of the clusters.

Publication types

  • Review

MeSH terms

  • Algorithms
  • Cluster Analysis*
  • Computational Biology / methods*
  • Neural Networks, Computer
  • Pattern Recognition, Automated