Next-Generation Cyber Threat Detection and Mitigation Strategies: A Focus on Artificial Intelligence and Machine Learning

Abstract

The principal objective of this research was to examine strategies for detecting and mitigating cyber threats in the next generation, by underscoring Artificial Intelligence (AI) and Machine Learning (ML). This study provides a comprehensive overview of the role of AI, ML, and deep learning (DL) in the domain of cybersecurity. Furthermore, this study highlights the benefits of integrating deep learning into cybersecurity practices. The researcher explored the effectiveness of consolidating AI and ML techniques into the Feedzai security system to reinforce the detection of fraudulent activities. To validate the methodology, the investigator experimented by employing the supervised machine learning random forest algorithm on a dataset comprising historical transaction records in CSV format. The results of the research ascertained that by employing Feedzai's AI-based software combined with the random forest algorithms, future financial institutions can achieve real-time fraud detection and accurate identification of legitimate transactions. The Random Forest framework had the highest accuracy rate, at 83.94%. By contrast, the Naïve Bayes framework had an accuracy rate of 79.23%, and the KNN model had the lowest accuracy rate, of 78.74%. These results ascertained that the Random Forest system was the most effective for pinpointing cyber-attacks.