Deep learning based QoS prediction method of Web service

Abstract

The invention discloses a deep learning based QoS prediction method of a Web service. A deep learning prediction model is provided for predicting the QoS (Quality of Service). The method comprises the following steps that QoS attribute data is preprocessed by using a wavelet transformation method, an original sequence is decomposed and reconstructed to form low-frequency and high-frequency sequences, deep circulation type neural network models are trained for subsequences, and in the prediction state, prediction values of the circulation type neural network models are added to obtain a final prediction value. To verify a prediction effect, QoS attributes including response time and throughput are predicted, and experimental results are compared via precision analysis and validity evaluation manners. Under time sequence samples with different features, the deep learning prediction model can maintain a higher prediction precision, is better than a single prediction model, and enables stable and effective prediction.

Description

technical field [0001] The invention relates to a method for predicting Web service QoS, in particular to a method for predicting service quality based on deep learning. Background technique [0002] Service-oriented systems increasingly access third-party Web services through the Internet, quality assurance and software maintenance are controlled by third parties, and the execution and management of the software itself also depends on third parties. The execution ability and service quality of service-oriented systems are increasingly dependent on services provided by third parties. However, in the complex and changeable Internet environment, this dependence on third-party services will bring uncertainty, making services unable to meet QoS (Quality of Service, quality of service) requirements. Therefore, it is necessary to predict the service quality, judge whether service failure may occur through prediction, and take actions in advance to eliminate or mitigate the negati...

AI Technical Summary

Problems solved by technology

Although Web service QoS prediction can improve the accuracy of Web service QoS, when the number of candidate Web services is large, it will take a certain amount of time to predict all Web services, and users often require that the optimal Web service can be selected as soon as possible. , how to select a model that can handle a large amount of data is another key issue

Images