• Rajeswari M
• Akanksha Mishra
• Alveena Pervaze
• K S Sumukha Vasista
• Pratham S

The escalating complexity of cyberattacks necessitates advanced, automated systems for real-time threat detection and mitigation. This research introduces an AI-driven cybersecurity framework implemented on a local high-performance workstation, utilizing machine learning to identify and counteract cyber threats. By integrating the CICIDS2017 and CSE-CIC-IDS2018 datasets with locally captured network traffic, the system employs a streamlined pipeline to preprocess data, fine-tune an XGBoost model, and achieve a 94.12% accuracy in multiclass attack classification, focusing on attack types in CSE-CIC-IDS2018. Threats are categorized into low, medium, high, and critical severity levels, enabling automated responses such as IP blocking and alerts for low-to-medium threats, while high-to-critical threats are flagged for human intervention. Comprehensive threat reports support informed decision-making. The system’s architecture, data processing, implementation, and evaluation are analyzed, highlighting its adaptability through local traffic integration. Performance metrics, including a 0.93 ROC-AUC, confirm its efficacy. Limitations include reliance on specific datasets and challenges with zero-day attacks. Future enhancements involve exploring ensemble models combining XGBoost and Random Forest. This research offers a scalable, adaptive solution for contemporary cybersecurity challenges.

"AI-Powered Cyber Threat Detection and Automated Response System", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 5, page no.i186-i202, May-2025, Available :http://www.jetir.org/papers/JETIR2505917.pdf

"AI-Powered Cyber Threat Detection and Automated Response System", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 5, page no. ppi186-i202, May-2025, Available at : http://www.jetir.org/papers/JETIR2505917.pdf