基于协同过滤的高校图书馆纸本资源智能推荐方法实证研究  

Empirical Research on Intelligent Recommendation Method of Printed Resources in University Libraries Based on Collaborative Filtering

作　　者：关芳[1,2] 高一弘 林强 Guan Fang;Gao Yihong;Lin Qiang(Xidian University Library,Xi’an Shaanxi 710126;Shaanxi Information Resource Research Center,Xi’an Shaanxi 710126)

机构地区：[1]西安电子科技大学图书馆,陕西西安710126 [2]陕西信息资源研究中心,陕西西安710126

出　　处：《情报探索》2020年第4期109-115,共7页Information Research

基　　金：中央基本科研业务费资助项目“面向科研项目全生命周期的知识管理与服务研究”(项目编号:JB180608)成果之一。

摘　　要：[目的/意义]旨在为高校图书馆提高纸质资源采购质量与利用率提供参考。[方法/过程]基于用户画像的理论对不同用户进行多维度的刻画,利用机器学习中监督学习的方法,通过采用协同过滤的推荐算法对用户偏好特征做精细统计分析的定量化计算,并从用户需求的角度建立用户偏好同步变化的自适应优化在线学习的纸本资源推荐系统。[结果/结论]该研究从实证分析角度为用户实现精准的个性化纸本资源推荐服务,为高校图书馆纸质文献检索库实现智能偏好的检索功能,建立纸质文献检索库合理有效的动态更新机制,提升用户体验。[Purpose/significance]The paper is to provide references for improving the quality and utilization of printed resources in university libraries.[Method/process]Based on the theory of user portraits multi-dimensional characterization of different users the paper uses the method of supervised learning in machine learning uses the collaborative filtering recommendation algorithm to quantitatively calculate the user’s preferred features and establishes an adaptive optimization online learning paper resource recommendation system that prefers synchronous changes from the perspective of user needs.[Result/conclusion]From the perspective of empirical analysis the paper realizes the accurate personalized paper resource recommendation service for users achieves the retrieval function of intelligent preference for paper document retrieval database of university libraries establishes a reasonable and effective dynamic update mechanism of paper document retrieval database and enhances the user experience.

关 键 词：协同过滤 高校图书馆 纸本资源 智能推荐 

分 类 号：G203[文化科学—传播学]