摘要
信息技术的广泛应用,导致网络信息快速增长,引发信息过载和混乱,使人们在搜索时难以找到有用数据,推荐系统的相关研究因此产生。基于传统搜索引擎技术的图书检索系统虽然具备高覆盖率,但无法准确反映读者偏好,导致图书推荐结果的准确率较低。针对上述问题,提出一种基于贝叶斯网络的分众化图书智能推荐算法。首先,利用关联规则算法对问卷调研数据进行挖掘,寻找读者与喜爱的图书类型之间的关联。其次,对关联规则算法挖掘出的结果进行深入分析。最后,利用贝叶斯网络建立一种个性化的图书智能推荐模型。实验结果表明,该方法提高了图书推荐结果的准确性与可靠度,且能够有效挖掘读者阅读偏好,简化图书推荐流程,达到了针对不同群体、不同种类的读者进行个性化图书推荐的目的。
The wide application of information technology leads to the rapid growth of network information,which triggers information overload and confusion,making it difficult for people to find useful data when searching,and the related research on recommender systems thus arises.Although the book retrieval system based on traditional search engine technology has high coverage,it cannot accurately reflect readers'preferences,resulting in low accuracy of book recommendation results.Aiming at the above problems,a Bayesian network-based intelligent recommendation algorithm for segmented books is proposed.First,the association rule algorithm is used to mine the questionnaire research data to find the association between readers and their favorite book types.Second,the results mined by the association rule algorithm are analyzed in depth.Finally,a personalized book intelligent recommendation model is established using Bayesian network.The experimental results show that the method improves the accuracy and reliability of book recommendation results,and can effectively mine readers'reading preferences,simplify the book recommendation process,and achieve the purpose of personalized book recommendation for different groups and types of readers.
作者
于超宇
窦水海
杜艳平
王兆华
YU Chaoyu;DOU Shuihai;DU Yanping;WANG Zhaohua(Beijing Institute of Graphic Communication,Beijing 102600,China)
出处
《北京印刷学院学报》
2024年第5期24-30,共7页
Journal of Beijing Institute of Graphic Communication
基金
北京市宣传文化高层次人才培养项目“融合视域下全民阅读分众化精准推广研究”成果。
关键词
贝叶斯网络
关联规则挖掘
分众化
图书智能推荐
Bayesian networks
association rule mining
audience demassification
intelligent book recommendations