Feature selection algorithms: a survey and experimental evaluation

TLDR

The proliferation of feature selection algorithms necessitates reliable criteria to determine the most suitable algorithm for a given situation. This study evaluates the performance of several fundamental feature selection algorithms in a controlled setting. The authors employ a scoring measure that ranks algorithms by relevance, irrelevance, and redundancy in sample datasets, assesses the match between algorithm output and the known optimal solution, and investigates the impact of sample size.

Abstract

In view of the substantial number of existing feature selection algorithms, the need arises to count on criteria that enables to adequately decide which algorithm to use in certain situations. This work assesses the performance of several fundamental algorithms found in the literature in a controlled scenario. A scoring measure ranks the algorithms by taking into account the amount of relevance, irrelevance and redundance on sample data sets. This measure computes the degree of matching between the output given by the algorithm and the known optimal solution. Sample size effects are also studied.