Hybrid neural networks for network security situation prediction: the CNN-LSTM-MulAttNet model

Yuxuan Li; Zihan Xiong; Jun Chen; Dabei Chen

doi:10.1117/12.3051995

15 January 2025 Hybrid neural networks for network security situation prediction: the CNN-LSTM-MulAttNet model

Yuxuan Li, Zihan Xiong, Jun Chen, Dabei Chen

Author Affiliations +

Proceedings Volume 13516, Fourth International Conference on Network Communication and Information Security (ICNCIS 2024); 1351619 (2025) https://doi.org/10.1117/12.3051995
Event: International Conference on Network Communication and Information Security (ICNCIS 2024), 2024, Hangzhou, China

Abstract

This paper proposed a hybrid Network Security Situation Prediction (NSSP) sequence neural network model, which integrates CNN, LSTM, and Multihead Attention mechanisms, and experimented with various techniques to enhance the nonlinearity and generalization ability of neural networks. We constructed NSSP sequence datasets on two publicly available anomaly detection datasets and validated the effectiveness and accuracy of our proposed method. The RMSE values of our proposed method on the testing datasets of UNSW-NB15-January 22, UNSW-NB15-February 17, and KDDCup99 are 2.262E-07, 2.275E-03, and 4.938E-05 respectively. The MAE values are 4.756E-04, 2.667E-02, and 3.322E-04 respectively.

(2025) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Yuxuan Li, Zihan Xiong, Jun Chen, and Dabei Chen "Hybrid neural networks for network security situation prediction: the CNN-LSTM-MulAttNet model", Proc. SPIE 13516, Fourth International Conference on Network Communication and Information Security (ICNCIS 2024), 1351619 (15 January 2025); https://doi.org/10.1117/12.3051995

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

;

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE