Network service credible attribute dynamic prediction method

Abstract

The invention relates to a network service credible attribute dynamic prediction method, and belongs to the technical field of computer software engineering. The method comprises the following steps:firstly, monitoring network service credible attributes, constructing credible attribute value tensors, decomposing the tensors by applying a weighted non-negative tensor decomposition method, and extracting time-related hidden characteristic factors; secondly, performing prediction of the next time period on the time-related implicit characteristic factor obtained after decomposition by applyingan exponential smoothing prediction method, and performing calculation estimation on a network service credible attribute value of the next time period; the method has the advantages that the time dynamic relationship of the network service credible attribute is utilized, and the prediction accuracy of the network service credible attribute is improved. The method does not depend on a specific credible attribute, is directly applied to a network service credible attribute monitoring result, and has very high flexibility and adaptability. A weighted non-negative tensor decomposition algorithm and an exponential smoothing algorithm are combined, so that the prediction effect of the method is improved.

Description

technical field [0001] The invention relates to a method for dynamically predicting trusted attributes of network services, which belongs to the technical field of computer software engineering. Background technique [0002] The service-oriented architecture is a distributed computing method, which gathers the computing power dispersed in various parts of the network to jointly perform a computing task, thereby reducing the construction cost and time overhead of the distributed system. Due to various reasons such as hardware, software, human operation, and network overload, the failure of network services is inevitable when performing services, which may greatly affect the user experience and suffer losses. The high-reliability requirements of network services require effective estimation and prediction of various attributes such as reliability, availability, and security, so that failure warnings or recovery measures can be taken in time. Especially with the rapid growth o...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to propose a method for dynamic prediction of trusted attributes of network services, which improves the existing methods for predicting trusted attributes of network services, so as to deal with the problem that the current trusted attribute prediction cannot effectively utilize the dynamic correlation of multiple attributes. In order to improve the prediction accuracy of trustworthy attributes, it can be applied to the trustworthiness monitoring and prediction tasks in different network service systems, and achieve the purpose of accurately predicting the trustworthy attributes of network services

Images