Significance of classification scores subsequent to feature selection

Authors:
Gobert N. Lee;Murk J. Bottema
Affiliations:
School of Informatics and Engineering, Flinders University, Adelaide SA 5001, Australia;School of Informatics and Engineering, Flinders University, Adelaide SA 5001, Australia
Venue:
Pattern Recognition Letters
Year:
2006

Citing 4
Cited 1

Genetic Algorithms in Search, Optimization and Machine Learning

Genetic Algorithms in Search, Optimization and Machine Learning
Multiple Comparison Methods for Means

SIAM Review
Feature Subset Selection Using a Genetic Algorithm

IEEE Intelligent Systems
Dimensionality reduction using genetic algorithms

IEEE Transactions on Evolutionary Computation

Background intensity independent texture features for assessing breast cancer risk in screening mammograms

Pattern Recognition Letters

Quantified Score

Hi-index	0.10

Visualization

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.