Accurate Criterion of Advanced Persistent trouble( APT) juggernauts is challenging due to lapping attack ways, evolving adversary geste, and the unshaped nature of Cyber trouble Intelligence( CTI) reports. While former work demonstrated the effectiveness of a crossbred machine knowledge and rule- rested approach for APT criterion, it reckoned on static, report- position analysis. This paper extends the crossbred frame by incorporating temporal correlation of CTI reports, criterion confidence estimation, and answerable intelligence labors to meliorate robustness and critic trust. The proposed extension correlates trouble reports across time windows using recreating pointers and contextual patterns, enabling more stable and harmonious criterion opinions. An explainability estate provides transparent defense through point position benefactions, rule- rested confirmation, and index terrain. Experimental results indicate that the extended frame reduces criterion nebulosity among lapping APT groups while maintaining computational effectiveness, making it suitable for deployment in real world security operations surroundings.

Keywords: Advanced Persistent Threats, Cyber Threat Intelligence, APT Attribution, Hybrid Machine Learning, Rule-Based Analysis, Temporal Threat Modeling, Explainable Artificial Intelligence(XAI), Attribution Confidence Calibration, Indicators of Compromise(IoCs), Threat Intelligence Analysis.