A NOVEL ADAPTIVE THREAT INTELLIGENCE FRAMEWORK FOR NEXT-GENERATION NETWORK SECURITY
DOI:
https://doi.org/10.62643/Keywords:
Threat intelligence, adaptive security, online learning, concept drift, STIX, TAXII, intrusion detection, automated response, network security.Abstract
The increasing complexity and frequency of cyber-attacks in next-generation networks require intelligent, autonomous, and adaptive defense mechanisms. Traditional intrusion detection and threat management systems often fail to respond effectively to zero-day attacks, advanced persistent threats (APTs), and rapidly evolving malware. To overcome these limitations, this study presents A Novel Adaptive Threat Intelligence Framework for Next-Generation Network Security, designed to deliver proactive, scalable, and context-aware cyber defense. The proposed framework introduces a four-layer adaptive threat intelligence architecture. First, a Multi-Source Threat Intelligence Collector gathers and normalizes data from network traffic logs, endpoint telemetry, vulnerability databases, and external threat feeds. Second, a Deep Hybrid Anomaly Detection Engine, combining Convolutional Neural Networks (CNN) and Bidirectional LSTM (BiLSTM), performs behavior-aware traffic analysis to accurately detect anomalies and unknown attack signatures. Third, a Threat Intelligence Fusion and Correlation Module integrates machine learning outputs with contextual threat indicators using a weighted fusion strategy to enhance detection confidence. Finally, a Reinforcement Learning–Based Automated Response Engine continuously learns optimal response actions, enabling dynamic mitigation, policy updates, and predictive threat management. an Adaptive Threat Intelligence Framework designed to strengthen next-generation network security through continuous learning, multi-source fusion, and automated response orchestration. ATIF-Net integrates heterogeneous telemetry (network flows, host logs, threat feeds, endpoint telemetry, and user behaviour), normalizes and enriches indicators using standardized schemas (STIX/TAXII), applies online and drift-aware machine learning models for detection and scoring, and drives automated, policy-guided responses via a decision orchestration engine.
Downloads
Published
Issue
Section
License

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.













