一种混合式协作过滤服务推荐算法  

A Service Recommendation Algorithm based on Hybrid Collaborative Filtering

作　　者：张圣[1] 

机构地区：[1]南京工业大学电子与信息科学学院,江苏南京210009

出　　处：《通信技术》2011年第7期118-119,122,共3页Communications Technology

基　　金：国家自然科学基金(批准号:71072172)

摘　　要：随着电子商务的飞速发展,服务推荐系统的重要性越来越得到体现。服务推荐技术作为电子商务的重要应用技术之一,为用户正确的决策提供有力的支持和保证。传统的基于协作过滤的服务推荐算法存在无法进行双向推荐的弊端,同时推荐质量有待提高。针对这些不足,提出了一种基于混合式协作过滤的服务推荐算法,考虑用户和服务之间的相似度,实现了双向推荐。实验结果表明,该算法能有效提高推荐质量。With the rapid development of e-commerce,service recommendation system becomes increasingly important.The service recommendation technology,as an important applied technology of e-commerce,could provide forceful support and guarantee the correct decision of the users.Traditional service recommendation algorithm based on collaborative filtering could not make bidirectional recommendation and the quality of recommendation system needs to be improved.To address this issue,a bidirectional service recommendation algorithm based on hybrid collaborative filtering is proposed.The similarities of users and service items are both taken into account,and the bidirectional recommendation is achieved.The experimental results show that this method could efficiently improve the service quality of recommendation system.

关 键 词：数据挖掘 混合式协作过滤 双向推荐 

分 类 号：TP391.3[自动化与计算机技术—计算机应用技术]