Embedding a 4:1 false-negative-to-false-positive penalty ratio into DRL reward signals reduces missed ransomware detections by 43% relative to symmetric rewards across 480 controlled runs.
Dynamic feature dataset for ransomware detection using machine learning algorithms
3 Pith papers cite this work, alongside 90 external citations. Polarity classification is still indexing.
3
Pith papers citing it
90
external citations · OpenAlex
fields
cs.CR 3years
2026 3representative citing papers
SISA training lets RL ransomware detectors forget selected samples by retraining one shard, with under 0.05% F1 drop and much lower retraining cost than full retraining.
RansomTrack hybrid framework detects ransomware at 96% accuracy in under 10 seconds via Radare2 static features, Frida dynamic behaviors, and ensemble ML on a public 165-family dataset.
citing papers explorer
-
SA-DRL: Security-Aware Deep Reinforcement Learning for Ransomware Detection with Asymmetric Reward Design
Embedding a 4:1 false-negative-to-false-positive penalty ratio into DRL reward signals reduces missed ransomware detections by 43% relative to symmetric rewards across 480 controlled runs.
-
Privacy-Aware Machine Unlearning with SISA for Reinforcement Learning-Based Ransomware Detection
SISA training lets RL ransomware detectors forget selected samples by retraining one shard, with under 0.05% F1 drop and much lower retraining cost than full retraining.
-
RansomTrack: A Hybrid Behavioral Analysis Framework for Ransomware Detection
RansomTrack hybrid framework detects ransomware at 96% accuracy in under 10 seconds via Radare2 static features, Frida dynamic behaviors, and ensemble ML on a public 165-family dataset.