A context-aware approach based Web service recommendation

Web services are the useful components, which are the integrated software widely used for the over the network machine-to-machine interaction.. It has got a great importance now-a-days. Hence, a recommendation system that recommends web services to service oriented application developers is vital. In this work, a personalized web service recommendation system is proposed to be developed which recommends web services to users by exploiting the location information of the user. The recommendations are customized to the users based upon the country in which they live by recommending services by considering the web services that are sensitive to their location. When a training user logs into the system they become the active user and the prediction is done for services that the particular user has not used yet. The global effects which are present in the user-service matrix are removed to avoid inaccuracies in predicting unknown values. An enhanced similarity measure is used to find similar items.