A fuzzy VSM-based approach for semantic service retrieval

Abstract

A vast number of business services have been published on the Web in an attempt to achieve cost reductions and satisfy user demand. Service retrieval consequently plays an important role, but unfortunately existing research focuses on crisp service retrieval techniques which are unsuitable for vague real world information. In this paper, we propose a new fuzzy service retrieval approach which consists of two modules: service annotation and service retrieval. Related service concepts for a given query are semantically retrieved, following which services that are annotated to those concepts are retrieved. The degree of retrieval of the retrieval module and the similarity between a service, a concept, and a query are fuzzy. Our experiment shows that the proposed approach performs better than a non-fuzzy approach on Recall measure.