Identifying the most influential scientific experts is of vital importance for exploring scientific collaborations to increase productivity by sharing and transferring knowledge within and across different research areas. However, most state-of-the-art expert finding approaches have usually studied candidates’ personal information and network information separately. In this dissertation research, we propose a Topical and Weighted Factor Graph (TWFG) model that simultaneously combines all the possible information in a unified way. In addition, we also design the Loopy Max-Product algorithm and related message-passing schedules to perform approximate inference on our cyclecontaining factor graph model. Information Retrieval will be chosen as the test field to identify representative scholars for different topics within this area. Finally, we will compare our approach with three baseline methods in terms of topic sensitivity, coverage rate of SIGIR PC members (e.g. Program Committees or Program Chairs) and NDCG (Normalized Discounted Cumulated Gain) scores for different rankings on each topic.