多维属性视角下学者用户画像构建及合作学者推荐研究  

Research on the Construction of Scholar User Portrait and the Recommendation of Cooperative Scholars from the Perspective of Multidimensional Attributes

作　　者：王大阜 邓志文[1] 贾志勇[1] 王静[1] Wang Dafu;Deng Zhiwen;Jia Zhiyong;Wang Jing(Library,China University of Mining and Technology,Xuzhou 221116,China)

机构地区：[1]中国矿业大学图书馆,江苏徐州221116

出　　处：《南京师大学报（自然科学版）》2023年第3期112-122,共11页Journal of Nanjing Normal University(Natural Science Edition)

基　　金：江苏省高校哲学社会科学研究项目(2022SJYB1129);国家社会科学基金项目(22BTQ023).

摘　　要：从多个维度属性构建学者用户画像,向目标学者推荐属性特征相似度高和易建立合作关系的合作学者,有助于增强学术交流合作、促进科研产出.本文以学者的论文成果为数据源,设计并探讨了合作学者推荐系统模型.首先采用Louvain社区发现算法划分学者社团,其次根据学者的基础属性、学术能力、研究兴趣、社交影响力四个维度构建用户画像,并从学者合作网络中获取合作关系强度、Katz相似性指标,最后根据候选学者的融合推荐评分实现合作学者推荐.通过构建学者用户画像,呈现学者全面的特征信息,赋予推荐结果可解释性.实证表明本文所提出的推荐模型具有良好的推荐效果,为目标学者的合作学者遴选提供了决策依据.Building user portraits of scholars from multiple dimension attributes and recommending collaborators who have high similarity in attribute characteristics and are easy to establish cooperative relationships to target scholars will help strengthen academic exchanges and cooperation,promote scientific research output.Taking the bibliographic data of scholars’papers as the data source,this paper designs and discusses the model of collaborative scholars’recommendation system.Firstly,Louvain algorithm is used for community discovery.Secondly,user portraits are constructed according to the four dimensions of scholars’basic attributes,academic ability,research interest,social influence,and indicators of cooperation relationship strength and Katz similarity indicator are obtained from the author’s cooperation network.Finally,the recommendation of cooperative scholars can be realized according to the fusion recommendation score of candidate scholars.By constructing a scholar user portrait,the comprehensive characteristic information of the scholar is presented,giving the recommendation results interpretability.The empirical results show that the recommendation model proposed in this paper has a good recommendation effect,which provides a decision-making basis for the selection of cooperative scholars.

关 键 词：用户画像 智慧图书馆 合作学者推荐 作者合作网络 

分 类 号：TP391.3[自动化与计算机技术—计算机应用技术]