Social media-based e-commerce consumer behavior prediction model in marketing strategy

Abstract

The rapid development of information technology has entered the era of network big data, online shopping for young people has become a fashion, and social media platforms have gathered a large amount of consumer purchase data. In this paper, for the current social media facing the problem of user consumption behavior prediction accuracy, data mining technology is referenced to analyze and predict user consumption behavior. The entropy weight method is used to segment e-commerce consumers based on RFM, and on this basis, the simple Bayesian method is used to model consumer behavior and construct an algorithm suitable for analyzing and predicting consumer behavior using social media data. Consumers are categorized into important value customers (7.21%), important development customers (18.76%), important retention customers (7.32%), general value customers (9.86%), general development customers (37.14%), and general retention customers (19.71%). The accuracy rate (ACC) for social media-based e-commerce consumer behavior is 84.92%, which allows for more accurate predictions. The study provides a scientific foundation for e-commerce platforms or enterprise decision-making, incubates emerging industries by analyzing big data, addresses major user needs, and becomes a new engine for promoting social progress.