Service Discovery Using Ontology Encoding Enhanced by Similarity of Information Content

Abstract

With the rapid development of web service standards and technology, the number of web services on Internet is increasing rapidly. Consequently, discovering the right service to meet a user's requirements quickly and accurately is crucial for the service community. Many web service discovery methods use web service models with semantic descriptions based on ontologies, allowing to apply logical reasoning to the discovery task. However, requiring logical reasoning can lead to sacrifices in efficiency of web services discovery. To address this problem, this paper proposes a combination of ontology encoding with the similarity of information content approach. We encode the concepts in the ontology in a binary encoding in order to improve the discovery efficiency and then we calculate the semantic similarity of information content between services. Validation of efficiency of the proposed approach is conducted through an experiment using the owls-tc2.0 as benchmark test set. The experimental results show that the proposed method not only can improve the efficiency of service discovery, but also can significantly improve the accuracy of service discovery compared with other discovery methods.