Exploiting User Feedback to Improve Semantic Web Service Discovery
State-of-the-art discovery of Semantic Web services is based on hybrid algorithms that combine semantic and syntactic matchmaking. These approaches are purely based on similarity measures between parameters of a service request and available service descriptions, which, however, fail to completely capture the actual functionality of the service or the quality of the results returned by it. On the other hand, with the advent of Web 2.0, active user participation and collaboration has become an increasingly popular trend. Users often rate or group relevant items, thus providing valuable information that can be taken into account to further improve the accuracy of search results. In this paper, we tackle this issue, by proposing a method that combines multiple matching criteria with user feedback to further improve the results of the matchmaker. We extend a previously proposed dominance-based approach for service discovery, and describe how user feedback is incorporated in the matchmaking process. We evaluate the performance of our approach using a publicly available collection of OWL-S services.
Published in 2009