A Research Of Web Service Discovery Method Based On QoS Driven

Service-oriented architecture is a component model,which encapsulates different functional units of application as servicing,enabling services built in various systems to interact in a unified and common way through an inter-service interface.Among them,Web service is an important way to realize service-oriented architecture,technology is a research hotspot in many fields such as integrated computer communication,application system design and implementation,network information security and so on.How to discover the user demand service in the massive service quickly and accurately is the key to research.This paper makes an in-depth study on the service and the systematic service of procuratorial work in China,puts forward to the method of discovering QoS data-driven procuratorial service.This method studies the service function description in the service description document and the service quality data in the service call log,and puts forward to the service clustering method based on the VSM model and the service discovery algorithm based on LSTM.First,for the service function description in the service description document,the method regards it as a special kind of structured text,introduces vector space model in the field of natural language processing,quantifies the description documents,and determines several services that are semantically similar to the users' need by similarity calculation.Secondly,for the service quality data in the service call log,the service quality evaluation system is constructed,and the comprehensive service quality index which can effectively reflect the prosecutor's service quality requirements.The index is obtained by selecting six evaluation indicators:message transmission time,message response time,error occurrence error,error code,user satisfaction and user subjectivity score.On that basis,providing users with the current time call service quality reference value,so as to meet the users'functional needs on the basis of the quality of service data to achieve optimal service discovery.For the comprehensive service quality index of timing,through the LSTM network to record the quality of service change trend information.Finally,subsequent experiments and verification show that the system can complete the credible service discovery,and have higher accuracy,accurate rate,recall rate and F value than other methods,which fully prove the feasibility and effectiveness of the method.These research contents provide some new idea and methods for the research in the field of service discovery,which is of great significance.