融合儿童成长信息的协同过滤推荐算法

浏览次数： 11

分享到:

摘要:

在互联网母婴领域中,由于育婴网络自身的特殊性,推荐算法不仅与用户以及项目的信息有关还与儿童的数据信息有关,而传统的用户相似度计算并未考虑儿童的数据信息.针对此问题,重新定义用户相似度计算方法,将儿童的数据信息通过加权融合的方法融入用户相似度计算中,并提出一种融合儿童成长信息的协同过滤算法,实验结果表明,该算法的准确率与召回率都优于传统算法,推荐系统的推荐质量也有所提高.

In the Internet maternal and child field,due to the particularity of the baby-care network itself,the recommendation algorithm is not only related to the user and the project information but also to the child's data information,while the traditional user similarity calculation does not consider the child's data information.Aiming at solving this problem,the user similarity calculation method is redefined,and the children's data information is integrated into the user similarity calculation by weighted fusion method.A collaborative filtering algorithm that integrates children's growth information is proposed.Experimental results show that both the accuracy rate and the recall rate are superior to the traditional algorithms,and the recommended quality of the recommended system is also improved.

作者:

刘行兵刘孝飞司思张震

Liu Xingbing;Liu Xiaofei;Si Si;Zhang Zhen(College of Computer and Information Engineering,Henan Normal University,Xinxiang 453007,China)

机构地区:

betway官方app 计算机与信息工程学院

出处:

《betway官方app 学报：自然科学版》 CAS 北大核心 2020年第4期7-11,共5页

基金:

国家自然科学基金(U1804164,61902112) 河南省教育厅自然科学项目(17A520039).

关键词:

协同过滤用户相似度母婴

collaborative filtering user similarity mother and baby