Use of Artificial Intelligence for Deep Learning Based Security Detection Systems: A Systematic Review of Techniques

This systematic review examines the use of artificial intelligence (AI) and deep learning technologies in security detection systems, with a particular emphasis on current practices, new developments, and their implications for the field of cybersecurity. The study focuses on novel AI applications employing deep learning techniques in intrusion detection, anomaly detection, and threat detection, including CNNs and RNNs. Using a systematic literature review approach, a synthesis of recent literature encompassing various security domains was performed. Findings highlight the adoption of more sophisticated deep learning techniques into security detection mechanisms, which surpass older methods in accuracy, flexibility, and computational efficiency. Still, model explainability, generalization, and privacy issues constitute some of the major unsolved problems. The study provides a synthesis of contemporary trends and evaluates the application of AI on security techniques to provide useful insights to address these gaps. The study suggests adopting more advanced techniques to mitigate bias and invasion of privacy within AI models, better addressing the issues of deep learning opacity, and refining adaptation for extensive systems.