The ways in which a hybrid mechanism jointly influences co-authorship evolution is clarified. The mechanisms are categorized into different groups and the contributions of different categorized mechanisms are computed. An improved meta-path based model called multirelations-based link prediction in heterogeneous bibliographic network is proposed. Experiments presented in Library and Information Science (LIS) show the co-authorship prediction accuracy is significantly improved with a hybrid mechanism denoted by a combination of predictors with weights.