Matching of social events and users: a two-way selection perspective

Recent years have witnessed a growing trend that offline social events are organized via online platforms. Along this line, large efforts have been devoted to recommending appropriate social events for users. However, most prior arts only pay attention to the selections of users, while the selections of events (organizers), which lead to the “two-way selection” process, are usually ignored. Intuitively, distinguishing the two-way selections in historical attendances can help us better understand the social event participation and decision making process in a holistic manner. To that end, in this paper, we propose a novel two-stage framework for social event participation analysis. To be specific, by adapting the classic Gale-Shapley algorithm for stable matching, we design utility functions for both users and event organizers, and then solve two layers of optimization tasks to estimate parameters, i.e., capturing user profiling for event selection, as well as event rules for attender selection. Experimental results on real-world data set validate that our method can effectively predict the event invitation and acceptance, compared with the combinations of one-way-selection baselines. This phenomenon clearly demonstrates the hypothesis that two-way selection process could better reflect the decision making of social event participation.