It is commonly observed that social events are organized through online social network services, and in this manner, there are vested interests in studying event-oriented social gathering through SNSs. The focal point of existing studies has been put on the analysis of event profiles or individual participation records. While there is the dynamic mutual influence among target clients through their social connections, the effect of dynamic mutual effect on the social gathering remains unknown. With that in mind, in this project, we build up a discriminant system, which permits to integrate the dynamic mutual dependence of potential event participants into the discrimination process. In particular, we plan the gathering focused occasion support issue as a two-arrange variation discriminant system to catch the clients’ profiles and in addition their latent social connections. he validation on real-world datasets show that our method can effectively predict the event cooperation with significant margin compared and a state of art baselines. This approves the hypothesis that dynamic mutual influence could assume an important part in the decision making process of social event participation. Also, we propose the network pruning technique to additionally enhance the efficiency of our technical framework. Finally, we provide a case study analysis to show the application of our framework for event plan design task.