MLPSO: A Filter Multi-label Feature Selection Based on Particle Swarm Optimization
Hamid Bayati, Mohammad Bagher Dowlatshahi, Mohsen Paniri
Abstract
Feature selection is one of the preprocessing steps in data mining and machine learning which aims to cope with redundant and irrelevant features through dimensionality reduction. In this paper, for multi-label classification, a novel filter approach using Particle Swarm Optimizer (PSO) is introduced. According to PSO, at first, a population of particles is generated and divided into two equal groups and compete in pairs, the winners are moved to the next iteration and the losers learn from the winners, and at the end of each iteration, the objective function for all the particles is computed. Finally, based on the best particle, the salient feature subset will be selected. The comparison results between the proposed method and six state-of-the-art multi-label feature selection methods on benchmark datasets using the ML-KNN classifier on various multi-label evaluation measures show the superiority of the proposed method.