摘要
为了全面分析用户兴趣数据,提升用户对推荐结果的满意度,提出一种基于协同过滤的学习资源推荐算法。分别将课程偏好、知识范围度以及教师偏好作为特征参数,结合用户的历史行为数据,对其进行全面提取。根据提取结果,以特征参数为基础,为学习资源构建属性标签,通过计算标签与用户兴趣特征的相似度,确定最终的推荐结果。测试结果表明,用户对设计算法推荐的图书资源、视频资源以及线上课程资源学习程度均高于对照组,也对资源推荐结果表现了较高的满意度。
In order to conduct a more comprehensive analysis of user interest data and improve user satisfaction with recommendation results, this paper proposes a learning resource recommendation algorithm based on collaborative filtering.Characteristic analysis, curriculum preference, knowledge scope and teacher preference are taken as characteristic parameters respectively, and combined with historical behavior data of users, comprehensive extraction is carried out. Based on the extraction results, attribute labels are constructed for learning resources on the basis of feature parameters, and the final recommendation results are determined by calculating the similarity between labels and users’ interest features. The test results show that the users’ learning degree of the book resources, video resources and online course resources recommended by the design algorithm was higher than that of the control group, and the users also showed higher satisfaction with the resource recommendation results.
作者
刘晓蒙
LIU Xiaomeng(School of Information Engineering,Zhengzhou Shengda University,Zhengzhou Henan 451191,China)
出处
《信息与电脑》
2023年第1期63-65,共3页
Information & Computer
关键词
协同过滤
学习资源
推荐算法
兴趣特征
属性标签
相似度
collaborative filtering
learning resources
recommendation algorithm
interest feature
attribute tag
similarity