An empirical analysis of dynamic network model of international trade by using enterprise sample simulation and improved ANN algorithm

Abstract

With the support of enterprise intelligence system, an intelligent modeling method based on synthesizing various data sources and complex metabolic networks is adopted to provide technical support for the practical application of complex large-scale dynamic models. Academic research on international trade networks relies to a large extent on network models based on macro-national data and static analysis of international trade patterns, which are usually based on charts for trade flows and economic globalization. This paper presents an improved neural network algorithm for analyzing the complex dynamic meta-network model of international trade. By using a BP neural network to adjust the weighted values of the best neural network model, the need for continuous adjustment of neural network values is reduced, which may be an effective means to improve the efficiency of intrusion testing applications. The simulation results demonstrate that the model has good explanatory power. The globalization of enterprise markets, as the main driver of international business activities, has a significant impact on the establishment and development of national or regional international trade networks. Therefore, there is a need to study international trade networks using data at the micro level. This paper suggests an interdisciplinary approach to the study of international trade networks and a micro-level study of international trade networks. The model based on complex meta-network dynamic model elements, introduce the dynamic network representation, and use temporary labels to define the network boundary characteristics.