A New Method for Clustering Analysis Based on Pseudo Parallel Genetic Algorithm

According to the fact that the conventional clustering algorithm based on the clustering criteria to initialize sensitive and easily to the local extreme, a new method of clustering analysis based on pseudo parallel genetic algorithm is designed. In the proposed PPGA-based clustering analysis method, each category is coded by real numbers, and some illegal chromosomes are repaired by identification and restoration of empty class. Two mutation operators including discrete random mutation operator and optimisation direction mutation operator are designed to balance the local convergence speed and the global convergence performance, which is then combined with migration strategy and insertion strategy. For the purpose of verification and illustration, the proposed PPGA clustering analysis method is compared with K-means clustering algorithm and standard genetic algorithms via a numerical simulation experiment. The experimental results show the feasibility and effectiveness of the new PPGA-based clustering analysis algorithm.