FP-ABC: Fuzzy-Pareto dominance driven artificial bee colony algorithm for many-objective software module clustering

Abstract

Recently, multi-objective evolutionary algorithms (MOEAs) gained wide attention to solve various search-based software engineering (SBSE) problems. The formulation of software module clustering problem (SMCP) as SBSE problem facilitates the application of many MOEAs. Even though MOEAs are being used successfully to solve the SMCPs, the performance of Pareto-dominance based MOEAs degrade if the number of objective functions is greater than three. To address the large and complex many-objective SMCPs (MaSMCPs), changes in MOEAs are essential. This paper proposes a Fuzzy-Pareto dominance driven Artificial bee colony (FP-ABC) to solve the MaSMCPs effectively and efficiently. In this contribution, fuzzy-Pareto dominance and two external archive concepts have been integrated into artificial bee colony (ABC) algorithm. The fuzzy-Pareto dominance improves the selection process of candidate solution and two external archives concept helps in balancing the convergence and diversity. To validate the supremacy of the proposed approach, a comparative study is performed with the existing many-objective optimization algorithms such as Two-Arch2, NSGA-III, MOEA/D, and IBEA, over seven real-world problems. The statistical analysis of the results indicates that the proposed approach outperforms existing approaches in terms of modularization quality (MQ), coupling, cohesion, and Inverted Generational Distance (IGD).