Feature selection using misclassification counts

Adil Bagirov, Andrew Yatsko, Andrew Stranieri, Herbert Jelinek

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

2 Scopus citations

Abstract

Dimensionality reduction of the problem space through detection and removal of variables, contributing little or not at all to classification, is able to relieve the computational load and instance acquisition effort, considering all the data attributes accessed each time around. The approach to feature selection in this paper is based on the concept of coherent accumulation of data about class centers with respect to coordinates of informative features. Ranking is done on the degree to which different variables exhibit random characteristics. The results are being verified using the Nearest Neighbor classifier. This also helps to address the feature irrelevance and redundancy, what ranking does not immediately decide. Additionally, feature ranking methods from different independent sources are called in for the direct comparison.

Original languageBritish English
Title of host publicationAusDM'11 - Conferences in Research and Practice in Information TechnologyConferences in Research and Practice in Information Technology
Pages51-62
Number of pages12
StatePublished - 2010
Event9th Australasian Data Mining Conference, AusDM 2011 - Ballarat, VIC, Australia
Duration: 1 Dec 20112 Dec 2011

Publication series

NameConferences in Research and Practice in Information Technology Series
Volume121
ISSN (Print)1445-1336

Conference

Conference9th Australasian Data Mining Conference, AusDM 2011
Country/TerritoryAustralia
CityBallarat, VIC
Period1/12/112/12/11

Keywords

  • Classification
  • Dimensionality reduction
  • Feature ranking
  • Feature selection
  • Optimization

Fingerprint

Dive into the research topics of 'Feature selection using misclassification counts'. Together they form a unique fingerprint.

Cite this