PSO and genetic algorithm optimization BP neural network contrast in collaboration innovation performance evaluation

Abstract

Strategic emerging industries takes a leading role in the economic and social situation and long-term development. Cooperative innovation is an important start for the cultivation and development of strategic emerging industries. The performance evaluation are the means to grasp the level of the performance level of the research, and it is an important tool for the scientific analysis of the collaborative innovation. This research is based on the research object of the coordination and innovation of industry, university and research in the major provinces and cities of China. Through constructing the performance evaluation system of the industry-university-research cooperative innovation, the evaluation model of the collaborative innovation performance has been built. Studying and comparing the effect of BP neural network based on genetic algorithm and PSO in the performance evaluation. To measure the level of collaborative innovation performance objectively and find out the problems of the strategic emerging industries and the existence of the collaborative innovation in the industry, to facilitate the government to introduce the corresponding policies and methods. Research shows the Particle Swarm Optimization(PSO) and Genetic Algorithm (GA) as two popular evolutionary algorithms both have global optimization features, they can make up the defect of Back Propagation(BP) neural network which is easily falling into minimum value. In this paper, PSO algorithm and genetic algorithm are simulated and the results are compared with each other and themselves. Compared with genetic algorithm, PSO algorithm proved it is easier to get the optimal value.