원문정보
초록
영어
Main challenges to the current semantic web service technologies are exponential continuous growth in the number of services on the Internet, syntax based discovery, lack of common agreed upon semantic service standards and heterogeneity of ontologies. In this paper, a service discovery approach independent of semantic service description models is proposed to solve the challenges of the current web service discovery. The idea is to combine principles from machine learning, data mining, statistical techniques and measures of semantic relatedness to make the semantic web service discovery process more intelligent, efficient and effective. The proposed approach exploits the use of semantic as well as syntactic information present within the service description profiles. Our approach is unique in terms of its application to any web service description language and the use of Omiotis measure of semantic relatedness for service discovery. The proposed approach has been implemented on OWL-S based service descriptions profiles and is able to find semantic relationship between the services which were otherwise discarded by the OWL-MX matchmaker. Empirical analysis shows that the proposed method out performs the TF - IDF based cosine similarity method, Jaccard method and LSA based approach.
목차
1. Introduction
2. Literature Review
2.1. Related Work
2.2. Semantic Relatedness and the Omiotis Measure
3. Proposed Methodology for Semantic Service Discovery
3.1. Overview
3.2. Hybrid OSD Algorithm
3.3. Hybrid Semantic Interpreter
3.4. Similarity Match Engine
4. Empirical Evaluation and Results
4.1. Dataset and Tools Used
4.2. Evaluation Measures for Service Discovery Approach
4.3. Discovering OWLS-S based Services using OSD Approach:
4.5. Discussion
5. Conclusion and Future Work
References