Data traffic classification method and device, equipment and medium

By integrating data traffic semantic features and label semantic features in data traffic classification and combining hidden layer state, the accuracy and reliability problems of data traffic classification in the prior art are solved, and more efficient data traffic classification is achieved.

CN119179946AActive Publication Date: 2024-12-24SHANDONG YUNHAI GUOCHUANG CLOUD COMPUTING EQUIP IND INNOVATION CENT CO LTD
View PDF 7 Cites 1 Cited by

Patent Information

Application Number
CN202411295984.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-14
Publication Date
2024-12-24
Estimated Expiration
2044-09-14

AI Technical Summary

Technical Problem

The existing data traffic classification methods have problems such as user privacy leakage and high computing power consumption, and machine learning methods may lead to bias in classification results.

Method used

By obtaining the semantic eigenvectors of the data traffic to be classified and the label semantic eigenvectors, using the self-attention mechanism to mine and fusion of the association relationship, and combining the hidden layer state for label prediction, improving the accuracy and reliability of data traffic classification.

Benefits of technology

A more label discriminant data traffic representation vector is realized, which reduces classification deviation and improves the accuracy and reliability of data traffic classification.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119179946A_ABST
    Figure CN119179946A_ABST
Patent Text Reader

Abstract

The invention discloses a data traffic classification method and device, equipment and a medium, and relates to the field of data classification, and the method comprises the steps: carrying out the semantic feature extraction of each traffic feature in to-be-classified data traffic, and determining a data traffic semantic feature vector based on each feature vector obtained after the semantic feature extraction; mining an association relationship between the data traffic semantic feature vector and the tag semantic feature vector based on a self-attention mechanism to obtain an association feature vector; and performing adaptive fusion on the data traffic semantic feature vector and the associated feature vector to obtain a data traffic representation vector, and determining the correlation between the data traffic representation vector and the hidden layer state based on a self-attention mechanism so as to process the data traffic representation vector to obtain a target data traffic vector, and furthermore, an output vector is determined in combination with the hidden layer state, so that a classification result containing a plurality of prediction labels of the to-be-classified data traffic is obtained through label prediction, and the accuracy of data traffic classification is improved.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Public opinion text classification method and system based on multi-label embedding, terminal and medium

    CN113987187A

  • Network traffic classification method based on attention mechanism

    CN114818867A

  • Web attack detection method based on gating Transform

    CN116527357A

  • ERNIECN-GRU-based speech step automatic identification method, system and device and medium

    CN116756605A

  • Network traffic anomaly determination method and device, electronic equipment and readable storage medium

    CN117176417A

Cited By

  • PCB manufacturing quality analysis method, device and equipment

    CN121505498A