Improving social recommender systems

Researchers have introduced a framework for social recommender systems that is intended to enhance recommendation accuracy. The proposed social recommendation framework is based on the advice-taking theory, integrating the relationship indicators between users and recommendation sources. A social recommender system based on this framework employs various mechanisms for capturing relationship information. It captures information required to track user consumption patterns, construct user profiles, and compare profiles. It also captures information needed to establish social networks and propagate links to form indirect links. The framework includes all available relationship indicators that can be selected based on the domain in which the system is deployed and efficiency considerations. Efficiency depends on three key factors, such as efforts required by users, effort required by administrators, and privacy concerns.