A New Fuzzy and Correlation Based Feature Selection Method for Multiclass Problems
Feature selection is one of the most important subjects in machine learning and pattern recognition. The main idea in feature selection algorithms is selecting a subset of features which does not include irrelevant and redundant features. In this paper a feature selection algorithm using genetic algorithm and fuzzy sets theory is proposed which we call fuzzy and correlation based feature selection - FCFS. We apply four fuzzy systems to obtain the fitness function in genetic algorithm. This filter method selects a low size feature subset so that the relevancy of each feature with the target is maximized and the redundancy among the selected features is minimized. Relevancy and redundancy are calculated based on Pearson’s correlation coefficient criterion. Several experiments are provided to demonstrate the effectiveness of the idea in terms of the classification accuracy and the number of selected features. Some statistical tests are also used to show the significant differences between the proposed method and the other methods.
Disclaimer/Regarding indexing issue:
We have provided the online access of all issues and papers to the indexing agencies (as given on journal web site). It’s depend on indexing agencies when, how and what manner they can index or not. Hence, we like to inform that on the basis of earlier indexing, we can’t predict the today or future indexing policy of third party (i.e. indexing agencies) as they have right to discontinue any journal at any time without prior information to the journal. So, please neither sends any question nor expects any answer from us on the behalf of third party i.e. indexing agencies.Hence, we will not issue any certificate or letter for indexing issue. Our role is just to provide the online access to them. So we do properly this and one can visit indexing agencies website to get the authentic information.