Abstract

In this study, a sophisticated cyberattack model was created for real-time DDoS & MITM attacks. Clouds, servers, big data, and networks are only a few of the modern infrastructures where cyber threats have proliferated. Threats to these platforms mean that private information may have been compromised within their applications. Cyber security experts could choose from a plethora of tried-and-true static models, but they all fell short when faced with dynamic attacks and other, more complicated threats. Furthermore, earlier cybersecurity models had problems with both high-ToC and target-attack scalability. Using machine learning methods, this study develops a sophisticated model for evaluating cyber risks. The recommended Lite RFO model was built using a top-notch Python programing utility. This architecture has been used to monitor cyberattacks on any network and locate the specific attack with a reduced Tree of Controls (ToC). There was a substantial improvement in performance metrics like detection rate (98.43%), accuracy (98.23%), sensitivity (96.89%), and recall (94.56%).