Security anomaly detection for enterprise management network based on attention mechanism
With the rapid growth of data volume in enterprise management systems and the continuous complexity of network architecture, traditional network security protection methods are no longer sufficient to fully address the security challenges. In response to the problems of insufficient accuracy and hig...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2025-03-01
|
| Series: | Frontiers in Physics |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fphy.2025.1538605/full |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | With the rapid growth of data volume in enterprise management systems and the continuous complexity of network architecture, traditional network security protection methods are no longer sufficient to fully address the security challenges. In response to the problems of insufficient accuracy and high time consumption in traditional network security anomaly detection methods, this paper proposes a detection model combining attention mechanism based spatial convolutional network and gated attention transformer (AMSCN-GADetector). It is an enterprise management network security anomaly detection method based on deep learning, aiming to achieve efficient and intelligent monitoring and management of security anomaly data in enterprise management network. This method combines spatial convolutional network and gating mechanism, which are used to extract spatial features from enterprise management network security data and learn non-local interaction relationships between features. In addition, by introducing attention mechanism, AMSCN-GADetector can accurately calculate the importance weights of network security data features. This effectively reduces the loss of critical security information in the detection process. Finally, through comparative experiments, it is verified that AMSCN-GADetector exhibits superior detection performance compared to other models, providing solid technical support for the stable operation and long-term development of enterprise management. |
|---|---|
| ISSN: | 2296-424X |