中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
A collaborative filtering recommendation algorithm based on biclustering

文献类型:会议论文

作者Wang JS(汪家升); Song H(宋宏); Zhou XF(周晓锋)
出版日期2015
会议名称2015 IEEE International Conference on Cyber Technology in Automation, Control, and Intelligent Systems (CYBER)
会议日期June 8-12, 2015
会议地点Shenyang, China
关键词collaborative filtering recommender system biclustering data sparsity
页码803-807
中文摘要Collaborative filtering has been widely used in many fields such as movie recommendation and e-commerce. However, there are still some problems such as data sparsity which restrict its further development. To address the data sparsity problem we proposed a novel collaborative filtering recommendation algorithm based on biclustering. Firstly, we use biclustering algorithm simultaneous clustering of the rows and columns of the rating matrix to generate biclusters, then the missing data can be smoothed by using the information of the biclusters. Secondly, a weighted matrix is introduced to distinguish between the original data and the smoothing data. Lastly, the active user's neighbors can be found based on the new similarity we proposed, and the recommendation to the active user is produced. The experiment results are offered, which show that the algorithm we presented can alleviate the data sparsity problem and improve the quality of the recommendation.
收录类别EI ; CPCI(ISTP)
产权排序1
会议录2015 IEEE International Conference on Cyber Technology in Automation, Control, and Intelligent Systems (CYBER)
会议录出版者IEEE
会议录出版地Piscataway, NJ, USA
语种英语
ISSN号2379-7711
ISBN号978-1-4799-8730-6
WOS记录号WOS:000380502300153
源URL[http://ir.sia.cn/handle/173321/17356]  
专题沈阳自动化研究所_数字工厂研究室
推荐引用方式
GB/T 7714
Wang JS,Song H,Zhou XF. A collaborative filtering recommendation algorithm based on biclustering[C]. 见:2015 IEEE International Conference on Cyber Technology in Automation, Control, and Intelligent Systems (CYBER). Shenyang, China. June 8-12, 2015.

入库方式: OAI收割

来源:沈阳自动化研究所

浏览0
下载0
收藏0
其他版本

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。