An Improved Genetic Algorithm For Emotion Feature Selection

A new improved genetic algorithm for solving the problem of emotion feature selection is put forward.The improved algorithm uses a population diversity operator,which is designed from two-aspects gene and individual,to initialize population with better distribu-tion .As the population individual passed through operator of the genetic algorithm tends to get close to the best point, we apply to the tabu search for local search.The simulation results show that the proposed algorithm can extend the search scope and overcome premature convergence of the GA effectively, and it can be satisfied with the best results of feature subset.