The Improvement Of Link Prediction Algorithm And Its Application In Industrial Structure Adjustment

In recent few years,the path of economic development has been adjusted strategically in China,and economic growth has been slowed down.Its method for economic growth is focusing on growing steadily,restructuring,transferring and preventing risks.Among them,in the way of restructuring,adjusting the industrial structure uniformly and promoting development of each departments with uniformly and continually.Because of the wider cooperation in departments and the more freedom of circulation,China’s industrial departments has evolved from a chain structure to a network structure.Therefore,the network-based theoretical analysis of the optimization of industrial structure can better reflect the harmonization and sustainability of the optimization scheme.Therefore,it has great significance to use the theory of complex network to analyze the optimized problem of industrial structure.In this paper,the complex network theory and existed algorithm of link prediction are used to optimize the link prediction algorithm.This paper will propose a link prediction similarity algorithm based on design of experimental.Then it will be used to analyze the relationship and development status of China’s industrial departments,establish the network model and use the link prediction similarity algorithm to propose a reasonable plan to optimize the industrial path.The main research contents of this paper are as follows:(1)Improvement of link prediction algorithm based on design of experimental: The similarity indicators in the existed link prediction algorithm are mostly single indicators,and the indicator only suitable for a specific network structure.In this paper,a hybrid similarity index is proposed.It makes the weight of the mixed similarity index become more optimized by the method of experimental design.At the same time,it is used in the microblog social network and the metabolic network for empirical analysis.The results show that the new link prediction algorithm has higher accuracy and better applicability.(2)The analyzation of optimize industrial path based on new link prediction algorithm: modeling network based on input and output data of main-function region,using relevant network structure indicators to analyze the develop status of industrial network and the importance of each department;Subsequently,the hybrid similarity algorithm is used to analyze the industrial network structure of the Beijing-Tianjin-Hebei region and the Yangtze River Delta region,then this papers propose the relevant industrial path optimization strategies for the adjustment of industrial structure.