Concept Learning and Feature Selection Based on Square-Error Clustering

Mirkin, Boris G.

doi:10.1023/A:1007567018844

Concept Learning and Feature Selection Based on Square-Error Clustering

Boris G. Mirkin

MLJ 1999 pp. 25-39

doi:10.1023/A:1007567018844 /mlj/1999/mirkin1999mlj-concept/

Abstract

Based on a reinterpretation of the square-error criterion for classical clustering, a "separate-and-conquer" version of K-Means clustering is presented and a contribution weight is determined for each variable of every cluster. The weight is used to produce conjunctive concepts that describe clusters and to reduce or transform the variable (feature) space.

PDF MLJ Semantic Scholar

Cite

Text

Mirkin. "Concept Learning and Feature Selection Based on Square-Error Clustering." Machine Learning, 1999. doi:10.1023/A:1007567018844

Markdown

[Mirkin. "Concept Learning and Feature Selection Based on Square-Error Clustering." Machine Learning, 1999.](https://mlanthology.org/mlj/1999/mirkin1999mlj-concept/) doi:10.1023/A:1007567018844

BibTeX

@article{mirkin1999mlj-concept,
  title     = {{Concept Learning and Feature Selection Based on Square-Error Clustering}},
  author    = {Mirkin, Boris G.},
  journal   = {Machine Learning},
  year      = {1999},
  pages     = {25-39},
  doi       = {10.1023/A:1007567018844},
  volume    = {35},
  url       = {https://mlanthology.org/mlj/1999/mirkin1999mlj-concept/}
}