Abstract
The database diversity and heterogeneity render the information discovery process a difficult and complex task. Vocabulary problems can make search results from traditional search engines irrelevant to users. For effective research, better knowledge of user behavior is important because users can participate in research design and construction. User profile has an important role in Web service selection, as it gives us a general overview of Web service requester. It is used in search engines so as to help to deliver with appropriate information to user's preferences, interests, or profile. In this paper, we present two points, namely (1) multidimensional user profile construction and (2) benefit of information stored in the user profile for Web service extraction, applied in social networks. The experiment shows that our method can recommend the needed Web services in a faster time and with significantly precision.