Feature selection based on decision boundaries

Chulhee Lee, David A. Landgrebe

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

7 Citations (Scopus)

Abstract

A novel approach to feature selection for classification is proposed based directly on decision boundaries. It is shown how discriminantly redundant features and discriminantly informative features are related to decision boundaries. It is noted that only a portion of the decision boundary is effective in discriminating between classes. A procedure to extract discriminantly informative features based on a decision boundary is proposed. The proposed feature selection algorithm predicts the minimum number of features necessary to achieve the same classification accuracy as in the original space, and it finds the necessary feature vectors. Experiments show that the performance of the algorithm compares favorably with that of previous algorithms.

Original languageEnglish
Title of host publicationDigest - International Geoscience and Remote Sensing Symposium (IGARSS)
Editors Anon
PublisherPubl by IEEE
Pages1471-1474
Number of pages4
ISBN (Print)0879426756
Publication statusPublished - 1991 Dec 1
Event1991 International Geoscience and Remote Sensing Symposium - IGARSS'91 - Espoo, Finl
Duration: 1991 Jun 31991 Jun 6

Publication series

NameDigest - International Geoscience and Remote Sensing Symposium (IGARSS)
Volume3

Other

Other1991 International Geoscience and Remote Sensing Symposium - IGARSS'91
CityEspoo, Finl
Period91/6/391/6/6

All Science Journal Classification (ASJC) codes

  • Computer Science Applications
  • Earth and Planetary Sciences(all)

Fingerprint Dive into the research topics of 'Feature selection based on decision boundaries'. Together they form a unique fingerprint.

  • Cite this

    Lee, C., & Landgrebe, D. A. (1991). Feature selection based on decision boundaries. In Anon (Ed.), Digest - International Geoscience and Remote Sensing Symposium (IGARSS) (pp. 1471-1474). (Digest - International Geoscience and Remote Sensing Symposium (IGARSS); Vol. 3). Publ by IEEE.