一种基于协同过滤及语义相关度的Web服务推荐系统

SOA is playing a more and more important role in software development. The demand for web services is growing so fast that the traditional UDDI which is built for service index can't satisfy developers. Software engineers need a more efficient way to find appropriate web services with which they can build more sophisticated applications. This paper introduces a new web service recommendation algorithm which is based on collaborative filtering and semantic relatedness. Cold start problems can be resolved when applying semantic relatedness into the system. The weights of both algorithms vary according to the user preference data size in order to keep the recommendations fit. The system can give good recommendations and help engineers to use web services efficiently.