Study on fairness of group recommendation based on stochastic model checking

With the increasing development of social networks, group recommendation has gradually become a hot research topic in the field of recommendation technology. The core of improving the quality of group recommendation lies in coordinating the preference differences among group members and improving the fairness among group members. The feedback loop formed by the user-system interaction can amplify the fairness bias in the system. In this paper, we address this common problem by formally modeling the general group recommendation framework into a DTMC model based on feedback loops and verifying fairness; we propose a dynamic fairness threshold detection framework based on stochastic model testing to detect the fairness of the system and make adjustments in time. The improved framework is simulated and tested, and its fairness is significantly improved.