| Web service becomes worldwide popular due to its loosely coupled architecture and the ease of developing it. In result, more and more individuals and companies design and develop their own web services and publish them online. This situation makes web service recognition an important issue in the field of web service research, which is also the key issue in this paper. There are two kinds of Web services:the traditional structured WSDL web service and the unstructured RESTful web service. Up to now, there are many mature recognition algorithms for the first one, whereas very few for the second.Against this background, this research concentrates on designing a distributed web services crawler engine, aiming to recognise and crawl both wsdl web services and RESTful web services efficiently and accurately.This paper first introduces some pertinent theoretical and technical knowledge about the distributed web services crawler engine, which helps determine the technical routemap of this research, followed by an analysis of both functional and non-functional requirements in the distributed web services crawler engine. Then this paper brings up an algorithm for RESTful web service recognition based on imporved naive bayes categorization algorithm, with which comes along theoretical analysis and experimental verification. Experimental results indicate that the algorithm has good performance in precision and recall and therefore fulfills the requirements for recognizing RESTful web services. The following are an introduction to the architecture of distributed web services crawler engine, a close illustration of the interaction flow in and among key modules, and a detailed explanation of the design of key modules. After that, this paper elaborates on the implementation of a prototype system and the its testing, which shows that the prototype system can fulfill the requirements. In the end, there are summaries and expectation of this research. |
PDF Full Text Request