An Improved Pseudo-parallel Genetic Algorithm for Process Mining

摘要

Process mining is helpful for deploying new business processes as well as auditing, analyzing and improving the already enacted ones. An improved pseudo-parallel genetic algorithm is proposed with an asexual reproduction for avoiding crossover operators' breach to nice gene patterns. The initial population is produced by greedy algorithm in order to enhance convergence velocity. Information exchange between subgroups employs island model in pseudo-parallel genetic algorithm. These measures are of great significance on reducing complexities and enhancing convergence velocity, as well as increasing global searching ability of the algorithm.