Consensus Feature Ranking in Datasets with Missing Values

Abstract

Development of a feature ranking method based upon the discriminative power of features and unbiased towards classifiers is of interest. We have studied a consensus feature ranking method, based on multiple classifiers, and have shown its superiority to well known statistical ranking methods. In a target environment such as a medical dataset, missing values and an unbalanced distribution of data must be taken into consideration in the ranking and evaluation phases in order to legitimately apply a feature ranking method. In a comparison study, a Performance Index (PI) is proposed that takes into account both the number of features and the number of samples involved in the classification.

DOI: 10.1109/ICMLA.2010.117

Extracted Key Phrases

Cite this paper

@article{Fakhraei2010ConsensusFR, title={Consensus Feature Ranking in Datasets with Missing Values}, author={Shobeir Fakhraei and Hamid Soltanian-Zadeh and Farshad Fotouhi and Kost V. Elisevich}, journal={2010 Ninth International Conference on Machine Learning and Applications}, year={2010}, pages={771-775} }