ThreatClassifier: Real-Time Cyber Attack Classification Using Hybrid ML Models
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
Real-time threat detection and classification is the core requirement for the cybersecurity discipline since the threat of advanced cyberattacks has increased multiple with the rapid expansion of digital technologies. Classic intrusion prevention and detection commonly utilize signature-based or single-model solutions that are unable to combat growing network traffic, new attack patterns and zero-day attack. This paper suggests ThreatClassifier, an online cyberattack classifier using hybrid machine learning models in a bid to reverse these downsides. The classifier combines the strengths of supervised learning models like Random Forest, XG-Boost, and Logistic Regression for maximum flexibility as well as accuracy in identifying multiple attack patterns. For efficient model optimization, feature engineering and dimension reduction are utilized, and ensemble learning methods provide immunity against spurious positives. With preventive security actionable insights, the intended system is aimed at detecting attacks like hacking, phishing, Distributed Denial of Service (DDoS), and malware infection in real-time. Also included in the system is a real-time visualization dashboard and user-friendly interface to allow security analysts to monitor threats, discover anomalies, and make timely decisions. Experimental evaluations on benchmark cybersecurity datasets demonstrate that the hybrid model achieves greater accuracy, faster classification, and improved cost- effectiveness compared to conventional models. By combining predictive potential and usability, ThreatClassifier increases proactive defenseso that organizations can remain strong against emerging cyber threats.
Fields of Study & Subjects
Network Security and Intrusion Detection
Internet Traffic Analysis and Secure E-voting
Network Packet Processing and Optimization
Publication details
- Journal / source
- International Conference on Intelligent Data Communication Technologies and Internet of Things (IDCIoT)
- Pages
- 1–6
- Publication date
- 4 February 2026
- Language
- EN