Leveraging fuzzy dominance relationship and machine learning for hybrid web service discovery

Nowadays, web service discovery plays an important role in several application domains. Existing semantic web service matchmakers usually use crisp logic-based matching, token-based similarity measures, and eventually machine learning that combines the individual scores into a global score. Unfortunately, these approaches entail an undesirable compensation between partial scores and consequently the final decision can be erroneous. Furthermore, there is no ideal matching similarity for assessing the parameters closeness, therefore, several similarity measures must be used to resolve the discovery issue. In this paper, we propose a hybrid semantic matchmaker that combines four textual similarity measures and a pure logic matching algorithm in order to search and rank the advertised services. The different scores are aggregated according to the fuzzy dominance function. We have conducted an exhaustive study based on the OWLS-TC benchmark to verify the effectiveness and efficiency of our approach.