Network traffic prediction methods, devices, electronic equipment and storage media

By combining deep forest models and temporal feature extraction networks, this approach solves the problem of the inability to effectively extract network traffic features in existing technologies, achieving higher prediction accuracy and generalization, and adapting to complex and dynamic network environments.

CN118353797BActive Publication Date: 2026-06-30INSPUR TIANYUAN COMM INFORMATION SYST CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
INSPUR TIANYUAN COMM INFORMATION SYST CO LTD
Filing Date
2024-03-26
Publication Date
2026-06-30

AI Technical Summary

Technical Problem

Existing technologies cannot effectively extract and select features that are helpful for prediction in network traffic forecasting, especially in high-dimensional data scenarios, resulting in insufficient performance of the model when dealing with complex patterns and hidden relationships.

Method used

A combined approach of deep forest model and temporal feature extraction network is adopted. Through multi-layer feature transformation and weighted averaging, combined with temporal prediction model and deep forest model, complex patterns and time series information are captured to improve prediction accuracy.

Benefits of technology

It improves the generalization and accuracy of network traffic prediction, enabling it to better extract valuable features from large amounts of user data, adapt to changing network environments, and achieve more accurate communication traffic prediction.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118353797B_ABST
    Figure CN118353797B_ABST
Patent Text Reader

Abstract

This invention provides a network traffic prediction method, apparatus, electronic device, and storage medium, belonging to the field of machine learning technology. The method includes: inputting time-series network traffic data into a first deep forest model to obtain a first prediction result output by the first deep forest model; inputting time-series network traffic data into a time-series prediction model's time-series feature extraction network to obtain a feature map output by the time-series feature extraction network; inputting the feature map into a second deep forest model to obtain a second prediction result output by the second deep forest model; inputting the feature map into a prediction network of the time-series prediction model to obtain a third prediction result output by the prediction network; and inputting the first, second, and third prediction results into a weighted averaging module to obtain a network traffic prediction result output by the weighted averaging module. This invention can extract valuable features from large amounts of user data and accurately predict communication traffic in a variable network environment.
Need to check novelty before this filing date? Find Prior Art