基于用户画像的高校图书馆个性化图书推荐研究
- 分享到:
摘要：
个性化推荐服务是高校智慧图书馆的建设重点,基于此,提出了图书推荐系统整体架构.首先从读者的属性、行为、兴趣等标签维度构建用户画像模型,其次考虑读者认知能力存在差异化的特点,将读者按照不同的身份类型划分,再结合基于协同过滤、内容及属性相似度的混合推荐算法进行图书推荐.最后,通过Hadoop大数据平台向目标读者推荐TOP-N图书,实验结果表明,基于该架构模型的图书推荐系统的推荐准确度高,并且有效缓解了推荐系统的冷启动问题.
Personalized recommendation service is the focus of the construction of University Smart Library.This paper puts forward the overall architecture of book recommendation system.Firstly,the user portrait model is constructed from the tag dimensions of readers'attributes,behavior and interests.Secondly,considering the characteristics of differences in readers'cognitive ability,the readers are divided into different identity types,and then combined with collaborative filtering a hybrid recommendation algorithm based on content and attribute similarity is used for book recommendation.Finally,Top-N books are recommended to target readers through Hadoop big data platform.The experimental results show that the book recommendation system based on this architecture model has high recommendation accuracy and effectively alleviates the cold start problem of the recommendation system.
作者：
王大阜 邓志文 贾志勇 安计勇
Wang Dafu;Deng Zhiwen;Jia Zhiyong;An Jiyong(Library,China University of Mining and Technology,Xuzhou 221116,China;School of Computer Science and Technology,China University of Mining and Technology,Xuzhou 221116,China)
机构地区：
中国矿业大学图书馆 中国矿业大学计算机科学与技术学院
出处：
《betway官方app 学报：自然科学版》 CAS 北大核心 2022年第3期95-103,共9页
Journal of Henan Normal University(Natural Science Edition)
基金：
江苏省高校哲学社会科学研究项目(2020SJA1009)。
关键词：
推荐系统 智慧图书馆 用户画像 冷启动
recommendation system smart library user profile cold start
分类号：
TP311.5 [自动化与计算机技术—计算机软件与理论]
