MVCWalker: Random Walk-Based Most Valuable Collaborators Recommendation Exploiting Academic Factors

摘要: In academia, scientific research achievements would be inconceivable without academic collaboration and cooperation among researchers. Previous studies have discovered that productive scholars tend to more collaborative. However, it is often difficult time-consuming for researchers find the most valuable collaborators (MVCs) from a large volume of big scholarly data. this paper, we present MVCWalker, an innovative method stands on shoulders random walk with restart (RWR) recommending scholars. Three factors, i.e., coauthor order, latest time, times collaboration, are exploited define link importance in social networks sake recommendation quality. We conducted extensive experiments DBLP data set order compare MVCWalker basic model RWR common neighbor-based friend friends various aspects, including, e.g., impact critical parameters factors. Our experimental results show incorporating above factors into can improve precision, recall rate, coverage rate recommendations.