| 评分偏差对于推荐质量的影响 |
| |
| 引用本文： | 胡必云,李舟军,王君,巢文涵.评分偏差对于推荐质量的影响[J].北京航空航天大学学报,2012,38(6):823-828. |
| |
| 作者姓名： | 胡必云 李舟军 王君 巢文涵 |
| |
| 作者单位： | 北京航空航天大学软件开发环境国家重点实验室,北京,100191;北京航空航天大学计算机学院,北京,100191;北京航空航天大学北京市网络技术重点实验室,北京,100191 |
| |
| 基金项目： | 国家自然科学基金资助项目，软件开发环境国家重点实验室自主研究课题资助项目 |
| |
| 摘 要： | 从理论上分析了评分偏差对于推荐质量的影响;基于潜在偏好及已知评分对评分偏差进行度量,其中潜在偏好通过心理测量学模型计算得出;通过设定不同的评分偏差水平,对评分偏差的影响进行了实验验证.理论分析及实验验证表明:评分偏差可导致推荐准确度及覆盖度下降;基于高质量的评分数据,协同过滤算法可为用户作出好的推荐.
|
| 关 键 词： | 人工智能 信号过滤与预测 信息检索 评分偏差 数据质量 协同过滤 推荐准确度 覆盖度 |
| 收稿时间： | 2011-03-18 |
Effect of rating residual on recommendation quality |
| |
| Hu Biyun Li ZhoujunState Key Laboratory of Software Development Environment,Beijing University of Aeronautics and Astronautics,Beijing,China Wang JunSchool of Computer Science and Technology,Beijing University of Aeronautics and Astronautics,Beijing,China Chao Wenhan.Effect of rating residual on recommendation quality[J].Journal of Beijing University of Aeronautics and Astronautics,2012,38(6):823-828. |
| |
| Authors: | Hu Biyun Li ZhoujunState Key Laboratory of Software Development Environment Beijing University of Aeronautics and Astronautics Beijing China Wang JunSchool of Computer Science and Technology Beijing University of Aeronautics and Astronautics Beijing China Chao Wenhan |
| |
| Institution: | 1. State Key Laboratory of Software Development Environment, Beijing University of Aeronautics and Astronautics, Beijing 100191, China;2. School of Computer Science and Technology, Beijing University of Aeronautics and Astronautics, Beijing 100191, China;3. Key Laboratory of Network Technology of Beijing, Beijing University of Aeronautics and Astronautics, Beijing 100191, China |
| |
| Abstract: | The effect of the rating residual on recommendation quality was analyzed.The rating residual was measured through user ratings and latent preferences.Latent preferences were computed with psychometric models.With different levels of rating residual,the effect of the rating residual was experimentally evaluated on real world datasets.Theoretical analysis and experimental results show that rating residual has negative effects on recommendation accuracy and coverage.Based on high quality of data,collaborative filtering algorithms can make precise recommendations for users. |
| |
| Keywords: | artificial intelligence signal filtering and prediction information retrieval rating residual data quality collaborative filtering recommendation accuracy coverage |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
| 点击此处可从《北京航空航天大学学报》浏览原始摘要信息 |
| 点击此处可从《北京航空航天大学学报》下载免费的PDF全文 |
