Service discovery and selection are commonly performed using either logic-based semantic matching techniques or other nonlogic-based approaches, such as data mining, linguistic processing, or information retrieval. This paper takes a hybrid approach for matching semantic Web services with users’ requests, using logic programming and syntactic similarity measurement. Essentially, the proposed solution “applies different matching filters to retrieve ... services that are semantically relevant to a given query,” in terms of seven matching degrees, aggregating their computed values to return a ranked list of services.

Although the actual problem that this paper attempts to solve is not stated clearly until the very end of the introduction, the paper is very didactic and readable. Readers who have a solid background in logic and semantic service ontologies will be best prepared to understand the foundations of the thoroughly explained solution.

This solution conveys a novel and interesting approach. The paper is accompanied by an implementation and evaluation analysis, where the suitability and performance of the proposal are discussed in detail. Furthermore, an early version of the implementation that provides the test cases showcased in the paper is available for download.

In conclusion, Klusch and Kaufer thoroughly explain and evaluate WSMO-MX, providing an excellent alternative to current discovery and selection approaches. However, the authors should have made available the actual implementation version they evaluated, so that interested readers could reproduce their experiments.