基于读者个性化特征的图书馆书目推荐

Library bibliographic recommendation based on personalized features of readers

采用协同过滤的图书馆书目推荐方法存在冷启动以及矩阵稀疏等问题,新书目未被读者评分或少量评分无法被推荐,存在推荐命中率低以及个性化服务差的缺陷。因此,提出基于读者个性化特征的图书馆书目推荐方法,设计基于读者个性化特征的数字图书馆书目推荐模型,给出图书馆书目个性化推荐流程,采集读者特征并分析读者行为后,塑造读者兴趣个性化特征,通过兴趣模型向读者推荐书目,塑造基于读者个性化特征的读者兴趣模型,基于显式信息和隐式信息对读者兴趣模型实时更新,进化读者个性化特征本体模型。分析基于读者个性化特征的图书馆书目推荐过程,个性化推荐业务实现过程。实验结果表明,所提方法可实现图书馆书目的个性化推荐,推荐准确率和效率较高。

The traditional library bibliographic recommendation method based on collaborative filtering has the problems of cold start and sparse matrix,low recommendation hit rate and poor personalized service because the new bibliography is not scored by readers or can′t be recommended with a small number of scores.Therefore,a library bibliographic recommendation method based on personalized feature of readers is proposed,a digital library bibliography recommendation model based on personalized feature of readers is designed,and the recommended procedure of library bibliographic recommendation is given.The reader feature is collected and the reader behavior is analyzed to construct the personalized features of interest of readers.The interest model is used to recommend books to readers,and the readers′interest model based on personalized feature of readers is constructed.On the basis of explicit information and implicit information,the readers′interest model is updated in real time to improve the ontology model of users′personalized feature.The library bibliographic recommendation process based on personalized feature of users and implementation process of personalized recommendation business are analyzed.The experimental results show that the proposed method can realize the personalized recommendation of library bibliography,and has high recommendation accuracy and efficiency.