PREDICTIVE CYBER ATTACK DETECTION IN CLOUD COMPUTING USING ARTIFICIAL INTELLIGENCE AND MACHINE LEARNING

Abstract

The scalability, adaptability, and cost-effectiveness of cloud computing have become it an indispensable element of contemporary digital infrastructure. The system's decentralized architecture renders it susceptible to data breaches, insider threats, zero-day vulnerabilities, distributed Denial of Service (DDoS) assaults, and several other cyber threats. Traditional intrusion detection systems struggle to identify novel and new threats in real time. This study introduces a novel approach for identifying cyber assaults in cloud systems via the use of machine learning techniques. The approach utilizes a hybrid deep learning architecture that integrates CNN and LSTM networks. The proposed method effectively identifies harmful activities by scrutinizing extensive network data, hence uncovering previously unrecognized attack patterns. Incorporating Explainable Artificial Intelligence (XAI) methodologies to elucidate model decisions improves transparency and fosters confidence. The proposed method provides a comprehensive and perceptive resolution to the issue of cloud computing security, as shown by experimental data demonstrating improved performance in recall, accuracy, precision, and false positive rate.