Proposal to Detect Cyber Assaults on Strategic Assets using Machine Learning

Abstract

The existing conventional approaches and strategies fails to address the complex, sophisticated, intelligent and fast adapting cyber assaults. Cyber Security (CySe) involves processes, the best practices, and technology to safeguard critical systems and networks from digital attacks. The paper addresses the principles, threats both conventional and contemporary and challenges for Cyber Security (CySe). Machine learning (MaLe) is the processes that allow computers to derive conclusions from data and enables the ability for computers to learn outside of their programming. Artificial Intelligence (ArIn) and computer science work together in machine learning (MaLe), where algorithms and data imitate human learning and improve accuracy over time. If leveraged well, it could simplify complex processes and incorporate more sophisticated and robust model for the cyber security. Machine learning techniques are required to improve the accuracy of predictive models. The paper evaluates the existing MaLe algorithms and proposes the model comprising of combination of Decision Tree and Random Forests to detect the cyber assaults on strategic assets. The paper highlights the advantages and challenges of MaLe algorithms and evaluates the proposed model using both qualitative and quantitative method in detecting cyber assaults.