S.Ilangovan1,*, A. Vincent Antony Kumar2
Intelligent Automation & Soft Computing, Vol.26, No.2, pp. 261-268, 2020, DOI:10.31209/2019.100000154
Abstract In this work, a Novel Feature selection framework called SU embedded PSO
Feature Selector has been proposed (SU-PSO) towards the selection of optimal
feature subset for the improvement of detection performance of classifiers. The
feature space ranking is done through the Symmetrical Uncertainty method.
Further, memetic operators of PSO include features and remove features are
used to choose relevant features and the best of best features are selected
using PSO. The proposed feature selector efficiently removes not only irrelevant
but also redundant features. Performance metric such as classification accuracy,
subset of features selected and running time are used for comparison. More >