An Adaptive Trust Prediction Framework for Diverse Data Model

The trust relationship is gaining importance in addressing information overload and personalized recommendation in rating based social networks. Current trust prediction models have significant drawbacks and limitations, particularly in terms of handling the diversity of user interactions. In this paper, we extract the features of trust relationship and classify trust metrics. Proposing a universality framework through comparing and qualitative analyze the existing prediction model. With this framework, we avoid the problem of the web of trust sparsity and solve the limitations of current methods when faced with diverse data models.