Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{192664,
        author = {Mohanakumari H S and Dr. Jithendra P R Nayak},
        title = {Smart Cybersecurity Framework: AI- Based Threat Detection Using machine and Deep Learning},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {9},
        pages = {1987-1992},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=192664},
        abstract = {Cyber threats are evolving rapidly, outpacing traditional security systems against zero-day exploits and APTs. This research proposes an AI-driven framework using machine learning (ML) and deep learning (DL) to automate real-time threat detection, prediction, and prevention. It addresses critical gaps like high false positives and slow response times by developing novel ML/DL algorithms for anomaly detection through behavioral analysis. A multi-layered defense architecture integrates supervised, unsupervised, and reinforcement learning, trained on datasets (e.g., CIC-IDS, NSL-KDD) to identify malware, phishing, and intrusions. Techniques include Graph Neural Networks (GNNs) for attack patterns, autoencoders for traffic anomalies, and Explainable AI (XAI) for transparency. The system employs adversarial training to resist evasion and federated learning for privacy-preserving authentication. Evaluations use precision/recall metrics, latency benchmarks, and adversarial stress tests, targeting a 50% reduction in false positives and sub-millisecond response times. Scalability is tested across cloud/edge environments, with lightweight models for IoT. Threat intelligence (e.g., MITRE ATT&CK) enables continuous retraining against APTs. The framework complies with GDPR/NIST standards and outperforms signature-based tools in simulations. Applications span finance, healthcare, and critical infrastructure. By bridging human expertise and autonomous AI, this research aims to redefine cyber defense paradigms. Findings will be shared via peer-reviewed publications and open-source prototypes.},
        keywords = {Artificial Intelligence (AI), Advanced Persistent Threat (APT) Mitigation, Behavioral Analysis in Cybersecurity, Adversarial Machine Learning, Real-Time Cyber Attack Prediction.},
        month = {February},
        }