Citation

Wei, Pin Chow and Chee, Yong Lau and Sek, Kit Teh and Chun, Lim Siow (2025) SVM-LSTM with SHAP for Dynamic API Security Whitelisting in AWS. In: 2025 Multimedia University Engineering Conference, MECON 2025, 21 July 2025 - 23 July 2025, Cyberjaya, Malaysia.

Text 480.pdf - Published Version Restricted to Repository staff only Download (3MB)

Abstract

Ensuring robust security for API Gateways is critical as they remain vulnerable to sophisticated cyber threats. This paper proposes a hybrid SVM-LSTM model to enhance anomaly detection for dynamic whitelisting and blacklisting in API security. The model leverages SVM for static classification and LSTM for temporal pattern recognition, significantly improving detection accuracy. Additionally, SHAP analysis enhances model interpretability by identifying key contributing features. Evaluated on publicly available CloudTrail logs, the proposed SVM-3LSTM model achieves 99.865% accuracy, 99.440% precision, 92.116% recall, and 95.638% F1-score, demonstrating superior performance in threat detection and mitigation. Notably, "eventname" was identified as the most influential feature, ensuring a more transparent and reliable API security system.

Item Type:	Conference or Workshop Item (Paper)
Uncontrolled Keywords:	SHAP; SVM; LSTM; blacklisting; AWS; API
Subjects:	Q Science > QA Mathematics > QA71-90 Instruments and machines
Divisions:	Faculty of Artificial Intelligence & Engineering (FAIE)
Depositing User:	Ms Suzilawati Abu Samah
Date Deposited:	19 Mar 2026 01:48
Last Modified:	19 Mar 2026 01:48
URII:	http://shdl.mmu.edu.my/id/eprint/15488

Downloads

Edit (login required)