Cloud Data Security and Intrusion Detection for DDoS Attack using Machine Learning
Bhavna Gangwar, Nupa Ram Chauhan · 2025
In recent advancement in cloud computing has become a significant challenge for the cloud data security as to increase un-ability to identify and mitigate any type of data security consent and risk of intrusion detection. Attackers continuously attack on cloud environment significant increasing the potential devastating consequences. With Artificial Intelligence techniques using machine learning are adaptable for the identification and prevention of vulnerability in data breach in cloud network by various attacks. The most important attack Denial-of-Services (DDoS Attack) continuously impact users and ISP (Internet Service Provider) because of its decentralized nature. This paper has proposed the supervised machine learning techniques for detecting network attack, intrusion detection and vulnerabilities in cloud environment. The proposed model, were using CIC-DDoS2019 dataset for identifying and predicting DDoS attack using machine learning as Random Forest, Logistic Regression, and Neural Network and KNN. These models have got the significant outcome in terms of Accuracy, F1-Score, Recall and Precision. The model has done data preprocessing, and feature selection and fine, tuning with hyper-parameter with the help of Bayesian optimization to obtain the best performance in the result of Random Forest to get the accuracy 99.95%, F1-score is 99.95%, precision is 100% and recall is 99.90%. This result show in the form of ROC-Curve, true positive rate and false positive rate.