| 利用云模型改进基于项目的协同过滤推荐算法 |
| |
| 引用本文： | 张新香,刘腾红.利用云模型改进基于项目的协同过滤推荐算法[J].图书情报工作,2009,53(1):117-117. |
| |
| 作者姓名： | 张新香 刘腾红 |
| |
| 作者单位： | 武汉中南财经政法大学信息学院 |
| |
| 摘 要： | 基于项目的协同过滤推荐算法能有效解决传统的基于用户的协同过滤推荐系统可扩展性差、缺乏稳定性的缺点，但仍然不能解决数据稀疏的问题，在数据极度稀疏的情况下，传统的项目相似性度量方法无法实现准确度量，导致推荐效果急剧下滑。本文借鉴基于云模型的云相似性度量方法来实现基于知识层面的项目相似性度量，提出了一种新的基于项目的协同过滤推荐算法。实验结果表明即使在数据极度稀疏的情况下，改进后的算法仍然能取得较好的推荐效果。
|
| 关 键 词： | 协同过滤推荐算法 云模型 相似度 |
| 收稿时间： | 2008-05-19 |
Applying Cloud Model to Improve Item-Based Collaborative Filtering Recommendation Algorithm |
| |
| Zhang Xinxiang,Liu Tenghong.Applying Cloud Model to Improve Item-Based Collaborative Filtering Recommendation Algorithm[J].Library and Information Service,2009,53(1):117-117. |
| |
| Authors: | Zhang Xinxiang Liu Tenghong |
| |
| Abstract: | Item-based collaborative filtering (CF) algorithms can deal with the scalability problems associated with used-based CF approaches without sacrificing recommendation or prediction accuracy. However, item-based CF algorithms still suffer from the data sparsity problems, the measuring method of items’ similarity works poor because of the extreme sparsity of user rating data, makes the quality of recommendation system decreased dramatically. This paper proposed a novel similarity measuring method on knowledge level, its thoughts comes from the measurement method of cloud similarity, then, based on the novel method, this paper puts forwards an improved item-based CF algorithms. Experiments results show that the algorithms can achieve better prediction accuracy even with extremely sparsity of data. |
| |
| Keywords: | collaborative filtering recommendation algorithm cloud model |
| 本文献已被 万方数据 等数据库收录！ |
| 点击此处可从《图书情报工作》浏览原始摘要信息 |
| 点击此处可从《图书情报工作》下载免费的PDF全文 |
