An evolutionary non-linear ranking algorithm for ranking scientific collaborations

摘要: The social capital theory motivates some researchers to apply link-based ranking algorithms (e.g. PageRank) compute the fitness level of a scholar for collaborating with other scholars on set skills. These are executed collaboration network and assign score each based scores his/her neighbors by solving linear system in an iterative way. In this paper, we propose new algorithm focusing link-aggregation function transition matrix. evolution strategy technique is applied find best aggregation matrix computing which modeled hypergraph. Experiments conducted two datasets gathered from ScivalExpert VIVO show that non-linear acts better than approaches scientific collaborations.