Ant Colony Optimization: A Review of Literature and Application in Feature Selection

Ant colony optimization (ACO) is a meta-heuristic that is inspired by real ants that are capable of exploring shortest paths, which inspires researchers to apply it for solving numerous optimization problems. Outstanding and acknowledged applications are derived from biologically activated algorithms like ACO that are established from artificial swarm intelligence which in turn is motivated by the amalgamated behavior of social insects. ACO is influenced by natural ants system, their behavior, team planning and organization, their integration for seeking and finding the optimal solution and also to preserve data of each ant. Currently, ACO has appeared as a popular meta-heuristic technique for finding the solution of conjunctional optimization problems that is beneficial for finding shortest paths via construction graphs. This paper highlights the behavior of ants and various ACO algorithms (their variants as well as hybrid approaches) that are used successfully for performing feature selection, applications of ACO and current trends. The fundamental ideas of ant colony optimization is reviewed including its biological background and application areas. This paper portrays how current literature utilizes the ACO approach for performing feature selection. By analyzing the literature, it can be concluded that ACO is a suitable approach for feature selection.