A network service quality prediction method and device

By combining Pre-Trained Transformer and graph neural network, and utilizing density clustering and sliding window techniques, a network service quality prediction method is constructed, which solves the problems of insufficient spatiotemporal correlation and data semantic utilization, and achieves higher accuracy prediction.

CN116743605BActive Publication Date: 2026-08-04PIO CLOUD COMPUTING (SHANGHAI) CO LTD
2 Cites 0 Cited by

Patent Information

Application Number
CN202310787386.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-06-29
Publication Date
2026-08-04
Estimated Expiration
2043-06-29

AI Technical Summary

Technical Problem

Existing network service quality prediction methods fail to effectively utilize spatiotemporal correlation and data semantic information. Furthermore, traditional methods are sensitive to anomalous data and have high sampling complexity, resulting in low prediction accuracy and network congestion.

Method used

By combining Pre-Trained Transformer and Graph Neural Network, a server distribution map is constructed based on geographical location. Core points are searched using density clustering, a set of linguistic descriptions of node features is generated, semantic embedding vectors are generated using sliding window, and a graph structure is constructed for feature extraction. Finally, a multilayer perceptron is used for prediction.

Benefits of technology

It improves the accuracy of network service quality prediction, reduces data redundancy and sampling complexity, reduces network congestion, enhances the utilization of spatiotemporal correlation and semantic information, and improves prediction accuracy.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The application discloses a network service quality prediction method and device, comprising the following steps: constructing a server distribution map according to a geographical position, searching for a core point by using a density clustering method; collecting QoS data of an edge cloud server corresponding to the core point, generating a language description set based on a Pre-Trained Transformer model; generating a characteristic word sequence of each edge cloud server based on the language description set and node characteristics, and generating a corresponding semantic embedding vector in a sliding window mode; constructing a graph structure based on the QoS data, and extracting features by using a graph convolutional neural network to obtain a topological feature vector; and splicing the topological feature vector and the semantic embedding vector, and predicting the network service quality of the edge cloud server by using a multilayer perceptron. The application integrates semantic information and space-time information to jointly complete the service quality prediction task, and improves the prediction accuracy.
Need to check novelty before this filing date? Find Prior Art