Cluster Analysis Research on Consumers’ Perceived Recommendation Trust

Abstract

Emerging social commerce relies on social media and is a new growth point for the development of e-commerce. However, it is difficult to cluster the recommendation information to reflect the subjectivity of consumers and the relationship between subjects. This paper constructs a clustering method based on consumer perception recommendation trust in the context of large-scale social networks and improves on subjective logic methods. Integrate trust characteristics into subjective logic trust transfer algorithm. Transform objective recommendation information into consumer subjective and differentiated perceived trust. Extract the similarity of perceptual recommendation trust and relationship intimacy from social networks to generate a normal matrix, and divide it by the method of spectral halving. A consumer perception trust network is extracted from social networks, a clustering method for consumer perception trust in social business is proposed from the perspective of complex network division, and a clustering center identification and update mechanism is designed for the high dynamic characteristics of social networks. The experimental results prove that: social business merchants and platforms quickly identify consumer perception of trust orientation, and provide methodological support for merchants to formulate trust-based precision marketing strategies for social business.