Study On The Discovery Method Of Semantic Web Service Based On Ontology

With the fast development of Internet, web applications base on Service-Oriented architecture become mainstream. There are a number of web services with different types on the internet. Effectively locating the necessary web services become research hot focus. The present technology of web service discovery based on keyword and service categories can easily lead to low accuracy and recall. Currently, the research on web services discovery devoted to the semantic web service discovery.In the traditional method of web service discovery based on semantic distance, measuring semantic similarity only consider the path with the shortest distance. However, the approach can't fully reflect the similarity between two concepts, when there are often have multiple paths between two concepts in the same domain ontology. In this paper, a novel method for computing concept similarity is proposed to solve the problem based on the weighted semantic distance. The method consists of two steps:finding out all paths between two concepts and measuring their semantic similarity. We first design an algorithm to search presently shortest past iteratively, and then let it find out all pasts between two concepts; secondly we calculate concept similarity using the weighted average distance of all paths, in the calculation of which we also consider the influences of the node depth and public parent simultaneously. We design a simulation experiment to validate the algorithm which we propose. The results show that the proposed method can improve the low accuracy and recall of web service discovery.In this paper, we outline the technology about semantic web service discovery based on ontology and domestic and foreign research status. We proposed a scheme to solve the shortage of the traditional algorithm of semantic web service discovery. This study provides a new approach which is benefit to effectively use web services on the internet.