Semantic-enriched data mining techniques for intensional service representation

The adoption of Web service technologies to enable collaboration in distributed environments has been made possible by the availability of huge amount of service repositories, that, if not properly controlled, leads to information overload rather than facilitating collaboration. Data mining provides well known exploratory techniques to extract relevant and frequent information from data repositories. This paper presents a preliminary effort to apply data mining algorithms to service repositories, to properly extract useful information about services. Our purpose is two-fold: (i) we study a proper Web service representation extracted from available Web service standards, to enable the application of data mining techniques; (ii) we propose the application of data mining algorithms to infer patterns representing summarized and integrated representation of service functionalities. These patterns can be used to facilitate the formulation of service requests on the underlying repositories. Semantic heterogeneities will be also addressed, in order to improve the recall of data mining results.