Measurement Index System and Empirical Study of Consumers' Continuous Trust on Social E-Commerce Platforms-Based on the GA-BP network

Abstract

Basing the research context on cross-border social e-commerce, this paper selects activity, online communication, social relationships, shared vision, content quality of merchants' posts and consumer purchase frequency to measure and predict consumers' continuous trust. The entropy weight method (EWM) is used to determine the weights, and the results show that two indices-social relationships and consumer purchase frequency - have a greater impact on the establishment of consumers' continuous trust. Based on the data from Wotobuy crawled using Python, this paper builds a consumers' continuous trust model based on the GA-BP neural network and conducts an error test. The results confirm the rather high accuracy of the model and suggest that it can be practically useful in the field of social e-commerce trust model.