Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
536899 | Pattern Recognition Letters | 2006 | 8 Pages |
Abstract
Feature selection prior to classification is shown to inflate performance. Empirical distributions are used to estimate statistical significance of classification scores. In an example study, eleven high classification scores are obtained but only three are found to be significant at p = 0.05.
Keywords
Related Topics
Physical Sciences and Engineering
Computer Science
Computer Vision and Pattern Recognition
Authors
Gobert N. Lee, Murk J. Bottema,