Issue No. 06 - November/December (2007 vol. 24)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/MS.2007.169
Aliaksandr Birukou , University of Trento, Italy
Enrico Blanzieri , University of Trento, Italy
Vincenzo D'Andrea , University of Trento, Italy
Paolo Giorgini , University of Trento, Italy
Natallia Kokash , University of Trento, Italy
Web service discovery is a difficult and challenging activity that makes the development of service-based applications a time-consuming and still not widely practiced process. Descriptions of publicly available services are scarce and their quality isn't guaranteed. This article presents a recommendation system to help developers of service-based applications discover and select appropriate services. Given a task description, the system recommends service operations according to the history of decisions previously made for similar objectives. The system is developed using IC-Service, a domain-independent recommendation Web service based on the implicit culture theory of service developers. IC-Service automatically collects information about service usage. Experimental results show that the system can learn from experience and achieve fair precision in its recommendations. This article is part of a special focus on service-centric software systems.
Web service discovery, recommendation systems, development of service-oriented systems
N. Kokash, A. Birukou, V. D'Andrea, E. Blanzieri and P. Giorgini, "Improving Web Service Discovery with Usage Data," in IEEE Software, vol. 24, no. , pp. 47-54, 2007.