Adaptive Cybersecurity Framework Utilizing Random Forest and SVM for Real-Time Threat Detection in Cloud Environments

Many cyber threats affect cloud computing environments such as zero day exploits and advanced persistent threats (APTs). In order to solve this problem, we put forward an adaptive cybersecurity model based on Random Forest and Support Vector Machines (SVM) for threat detection in realtime. This position is implemented in the current hybrid model of Random forest, which provides the feature set optimization from the Random forest while the SVM is able to handle the complexity of the data and the features and dimensions thereof. It is thus a system that can learn; the parameters of new threats are identified online hence enhancing new and better results constantly. As demonstrated by experimental data, the proposed framework accomplishes detection accuracy of 98.1%, precision 97.9%, and low false positives of 0.8%. The framework is modular, extendible and elastict, and can deal with large volumes of data in cloud environments, thus appropriate for real-time threat detection. This work advances the available literature on cloud security in an effort to offer a more adaptive and continuous method for countering cyber threats.