学术空间 / 论文 / 会议论文
a recommendation method for scholars based on trust and research interests
作 者 | saimei sun , , , , * |
会议名称 | the 15th web information systems and applications conference |
会议信息 | september 14-15 2018 taiyuan,shanxi,china |
发表日期 | 2018 年 09 月 |
摘 要 |
as the exponential growth of the number of users in academic social networks, it makes it difficult for users to find similar scholars whom users are interested in. thus, a recommendation method for scholars based on trust and research interests is proposed. firstly, we calculate the trust between users to measuring the valid relation strength by using the interaction behavior of the user in the social network. secondly, the three degrees of influence rule is utilized to expand the users’ potential friend set. thirdly, similarity of research interests among users is calculated by the lda topic model. finally, the trust and similarity are integrated as the comprehensive similarity by the weighted mean method to recommend users. the experimental results on datasets of scholat demonstrate that our method can effectively improve the performance of scholar recommendation in terms of precision and recall. |
关 键 字 | academic social networks; scholar recommendation; trust; three degrees of influence; lda |
基金项目 | 国家自然科学基金(61772211);广州市科技计划项目(201604046017);广州市科技计划项目(201704020203); |
附 件 |