Cyber Attack Detection System in University Private Cloud Using Machine Learning
M. S. Chethan, Channakrishnaraju Channakrishnaraju, R. Rajeswari, M. Selvam · 2023
As cloud computing adoption in colleges continues to rise, the security of private cloud systems has become a paramount concern. Data breaches resulting from cyber attacks can inflict severe damage to a university's revenue and reputation. This research proposes a novel machine learning-based cyber threat detection system tailored to the university's private cloud environment. The system's main objective is to continuously monitor the cloud infrastructure and employ advanced machine learning algorithms to analyze network traffic, identify and prevent unusual activities that may indicate potential cyber-attacks. Here, the challenges posed on two sides of known possible threats and attack worldwide come across, and administrative defaults leads to security hole. By leveraging the power of machine learning, this innovative system aims to enhance the university's cyber defence capabilities. It considers the dynamic and evolving nature of cyber threats, enabling real-time detection and proactive measures against malicious activities. The integration of cutting-edge machine learning models and feature extraction techniques empowers the system to identify patterns of anomalous behaviour, even in the face of sophisticated attacks. Key components of the proposed system include network traffic analysis, anomaly detection and threat intelligence integration. Through the analysis of network packets and access logs, the system can effectively detect signs of unauthorized access, data exhilaration, and other cyber threats. Additionally, threat intelligence feeds provide the system with up-to-date information on emerging threats, enabling quick responses to potential risks. Moreover, the system's implementation adheres to privacy and data protection regulations, ensuring secure handling of sensitive information within the private cloud environment. Regular updates and adaptive learning capabilities enable the system to evolve with changing cyber threats, ensuring continued robustness in the face of new challenges. In conclusion, the proposed machine learning-based cyber attack detection system presents a powerful solution to safeguard the university's private cloud infrastructure. By promptly detecting and mitigating potential cyber threats, the system acts as a proactive defence mechanism, safeguarding valuable data and preserving the university's reputation in the ever-evolving landscape of cyber security.