社区中基于QoS和相关度的Web服务推荐方法

An Approach for Web Service Recommendation in Community Based on QoS and Similarity

随着Web服务技术的发展,越来越多的社区引入了Web服务。为了更好地满足用户的要求,本文提出了一种为社区用户推荐服务的方法CWSR。CWSR首先对服务质量即QoS给出了定量的测算,然后通过解析服务的WSDL文档提取出服务关键字,计算社区与服务关键字的相似性从而获得服务对社区的相关度,最后结合两者与历史评分数据训练得出某特殊社区的推荐模型,根据该模型预测其它服务对于该社区的推荐指数。实验表明,本文提出的方法具有比以往方法更优的推荐质量。

With the development of Web service technology, more and more Web services are taken into practice in online communities. In order to better meet customers＇ requirements, a solution CWSR for recommending the optimal service to community users is proposed. First of all, the paper gives a quanti- tative measurement for the Quality of Service （QoS）,extracts services description from the WSDL docu- ment and calculates community and service keywords similarity. Finally, a web service recommendation model is proposed by combining the QoS and similarity with the history results. According to that model, the most appropriate service will be recommended to the appropriate users of community. Experiments show that our method makes the recommended services can better meet the needs of the community users.