| 基于群体兴趣偏向度的数字图书馆协同过滤技术研究* |
| |
| 引用本文： | 马丽.基于群体兴趣偏向度的数字图书馆协同过滤技术研究*[J].现代图书情报技术,2007,2(10):19-22. |
| |
| 作者姓名： | 马丽 |
| |
| 作者单位： | 西华师范大学商学院,南充,637002 |
| |
| 基金项目： | 西华师范大学校科研和教改项目 |
| |
| 摘 要： | 针对数字图书馆协同过滤推荐系统所面临的用户评分数据稀疏性问题，提出群体兴趣偏向度的计算方法，对用户－项目评分矩阵空缺值进行预测。实验结果表明，该算法能有效提高推荐质量。
|
| 关 键 词： | 群体兴趣偏向度 平均绝对偏差 |
| 收稿时间： | 2007-08-27 |
| 修稿时间： | 2007-08-27 |
Study on Digital Library Collaborative Filtering Technology Based on Group Interest Trend Degree |
| |
| Ma Li.Study on Digital Library Collaborative Filtering Technology Based on Group Interest Trend Degree[J].New Technology of Library and Information Service,2007,2(10):19-22. |
| |
| Authors: | Ma Li |
| |
| Institution: | Business College, China West Normal University, Nanchong 637002, China |
| |
| Abstract: | Collaborative filtering recommendation systems in digital library have faced the problem of sparse user ratings.To solve the problem,a computing method of group interest trend degree has been proposed and used into the prediction of vacant values in user-item matrix.The experimental results show that the algorithm can efficiently improve recommendation quality. |
| |
| Keywords: | Collaborative filtering Group interest trend degree Mean absolute error |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
| 点击此处可从《现代图书情报技术》浏览原始摘要信息 |
| 点击此处可从《现代图书情报技术》下载免费的PDF全文 |
