Research On Web Services Credibility Prediction Model Based On ACO-GRU

Web services have developed rapidly with their unique advantages of language independence,platform independence,low coupling,interoperability,and protocol standardization.More and more Web services with similar function but different quality have emerged on the Internet.When Web services are selected and recommended,the credibility of Web services in quality becomes an important indicator that user concerned.Therefore,effective measurement and prediction on Web service credibility have become an indispensable part of service applications.This paper carries out the research on Web service credibility,and defines and measures Web service credibility based on the quality of service attributes.Based on the analysis and summarization of relevant research results at home and abroad,this paper makes an in-depth study on Web services credibility and its prediction related technology theory.Firstly,A Web service credibility prediction model based on ACO-GRU(Ant Colony Optimization-Gated Recurrent Unit)is established.The model analyzes and predicts Web services credibility based on time series data,and mines the hidden rules in Web services credibility time series data to further predict the future credibility of the services.The main research contents of this paper are as follows:(1)GRU neural network is used to predict Web services credibility,which avoids the shortcomings of fully connected neural network in processing time series data,and improves the prediction accuracy on Web services credibility.(2)In order to solve the problem that GRU neural network is easy to fall into local optimal solution,this paper uses ant colony algorithm with global optimization ability to improve GRU network,and presents an initial weight threshold selection method for GRU networks based on ant colony algorithm.(3)The ACO-GRU-based Web service credibility prediction model is implemented.(4)Under the same experimental condition,the Web service credibility prediction algorithm studied in this paper is compared with time series prediction algorithms in other papers.Experiments show that the ACO-GRU-based Web service credibility prediction model has higher prediction accuracy.