Swarm Intelligence for Feature Selection: A Review of Literature and Reflection on Future Challenges

Abstract

Feature subset selection is considered to be a significant task in data mining. There is a need to develop optimal solutions with higher order of computational efficiency. In this paper, we reviewed the problems encountered during the process of feature selection and how swarm intelligence has been used for extraction of optimal set of features. It also gives a concise overview of various swarm intelligence algorithms like particle swarm optimization, ant colony optimization, bacteria foraging algorithms, bees algorithm, BAT algorithms and the various hybrid approaches that have been discovered using these approaches.