Solving graph partitioning problem using genetic algorithms

Abstract

The graph partitioning problem (GPP) is one of the fundamental multimodal, combinatorial problems that has many applications in computer science. Many deterministic algorithms have been devised to obtain a good solution for the GPP. This paper presents new techniques for discovering more than one solution to this problem using genetic algorithms. The techniques used are based upon applying niching methods to obtain multiple good solutions instead of only one solution. The paper also presents in detail a comparison between the results of a traditional method, simple genetic algorithm (SGA), and two niching methods, fitness sharing and deterministic crowding when applied to the graph partitioning problem.