A Hybrid Evolutionary Multi-objective Optimization Algorithm for QoS-driven Service Selection Problem

The QoS-driven Service Selection (QSS) problem is a wellknown NP-hard problem in the combinatorial optimization field. Although the QSS problem is naturally multi-objective optimization problem, most of the existing approaches solve the problem in single-objective optimization context. In the recent years, there have been some efforts to tackle the problem in multi-objective optimization context. In this paper, we propose a hybrid interactive Evolutionary Multi-objective Optimization (EMO) algorithm for solving the QSS problem in multi-objective context. The proposed algorithm hybridizes an interactive EMO algorithm with Path Relinking. The performance of the proposed algorithm is assessed for two and three objectives QSS problem sets. We show the performance advantage over other existing standard approaches such as NSGA-II. The comparative evaluations of results indicate that the proposed approach can converge to the preferred solution faster, and improve upon previously existing techniques up to 13.3% in terms of the requirement of user feedback.