A Cloud Intrusion Detection System Design by SMOTE Integration
K. V. K. Sasikanth, T. Sri Haripriya, K. Siva Rama Raju, J. Govind Raju, M. Praveen Kumar · 2025
By addressing class imbalance in network traffic datasets through the use of the Synthetic Minority Over-sampling Technique (SMOTE), this project offers an enhanced design for cloud intrusion detection systems (IDS). To improve detection skills, the study uses a large dataset that includes parameters like duration, protocol type, service, and packet statistics. We assessed well-known intrusion classification techniques such as Logistic Regression, Support Vector Machine, Decision Tree, and Random Forest, as well as suggested cutting-edge algorithms like AdaBoost, Stacking Classifier, and Naive Bayes. According to experimental data, our suggested algorithms perform better than conventional classifiers in terms of precision, recall, and F1-score, resulting in a cloud IDS framework that is more dependable. This study offers a fresh approach to the problems caused by unbalanced cybersecurity datasets.