Artificial Intelligence-Driven Threat Intelligence for Real-Time Cyber Attack Detection: A Hybrid Machine Learning Framework
Abstract
The growing complexity and frequency of cyber-attacks have exposed the limitations of conventional security systems that rely on signatures, static rules, and manual threat analysis. Artificial intelligence (AI)-driven threat intelligence has emerged as a promising solution for improving the speed, accuracy, and adaptability of cyber-attack detection in real time. This study investigates the role of machine learning and deep learning techniques in enhancing cyber threat detection and proposes a hybrid AI-based framework for real-time attack identification. A quantitative experimental design was adopted using cybersecurity datasets consisting of network traffic records, intrusion detection logs, and malware behavior traces. The proposed model combines Random Forest for structured feature-based classification and Long Short-Term Memory (LSTM) networks for temporal sequence learning. Performance was evaluated using accuracy, precision, recall, F1-score, false positive rate, and detection latency. The hybrid model outperformed traditional signature-based systems and single-model baselines, achieving an accuracy of 98.1%, recall of 97.4%, F1-score of 97.7%, and a false positive rate of 2.3%. The findings indicate that AI-driven threat intelligence significantly improves real-time detection capability, reduces false alarms, and supports proactive cyber defense. The study contributes a scalable and adaptable framework for intelligent cybersecurity operations across enterprise and cloud environments.
How to Cite
Sakshi Sharma, Mamta Sharma, "Artificial Intelligence-Driven Threat Intelligence for Real-Time Cyber Attack Detection: A Hybrid Machine Learning Framework", Vol. 3, Issue 8, 21-11-2025, pp. 34-45.
