Logo image
Back
Optimizing Cloud Security with CNN and XGBoost Models for Intrusion Detection Systems
Conference proceeding

Optimizing Cloud Security with CNN and XGBoost Models for Intrusion Detection Systems

Amit Karbhari Mogal, V. Anitha, Preeti Nitin Bhatt, Kriti Srivasatava, S. Rukmani Devi and Anvesh Perada
2025 3rd International Conference on Integrated Circuits and Communication Systems (ICICACS), pp 1-6
21 Feb 2025
DOI: https://doi.org/10.1109/ICICACS65178.2025.10968768

Abstract

Accuracy Computational modeling extreme gradient boosting (XGBoost) Feature extraction Integrated circuit modeling Intrusion detection intrusion detection system (IDS) Reliability Standardization Standards organizations Training Cloud Computing
Cloud computing and Internet growth have simplified certain formerly difficult activities. This advancement has also revealed many security weaknesses. Due to cyberattacks, organizations need Intrusion Detection Systems (IDS) to protect their data and networks. Preprocessing, feature selection, and model training comprise the suggested strategy. Preprocessing includes data cleansing, standardization, and labeling. Information gain, chi-square, and PSO are used to choose features. CNN-XGBoost trains the model. The CNN-XGBoost model outperforms solo CNN and XGBoost with an average accuracy of 92.08%. This accuracy proves the hybrid method's cloud computing breach detection usefulness. The study underlines the importance of enhanced IDS models in cyber risk reduction. CNN's feature extraction and XGBoost's classification power provide a powerful, efficient, and reliable cloud intrusion detection model. Modern organizations benefit from this network security method.

Metrics

1 Record Views

Details

Logo image