Fileless and AI-Powered Memory Malware Forensics: a Panoramic Detection Architecture With Machine Learning and Deep Learning Enhanced Volatility and Rekall Analysis

View at publisher Summarise with VIRAT
Authors: Gunaseelan C, N. Kamalraj
Cited by: 0
References: 15

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

Malware, or malicious software intended to disrupt, compromise data, or provide a barrier to authorised access, is increasingly taking a memory-resident and fileless form of execution, and as such, it bypasses older disk-based detection methods. In this study, a comprehensive malware detection and heuristic analysis framework utilizing Volatility and Rekall forensic toolsets is proposed to detect advanced inmemory threats. Since contemporary malware does not leave much disk traces, the framework adds volatile-memory analysis using multi-layered heuristic rules to interpret execution paths, maliciously behaving processes, memory evidence, and covert code injections. The synergistic nature of Volatility and Rekall allows them to extract, correlate, and verify memory indicators of stealthy runtime-persistent attacks with high accuracy. Moreover, opcode pattern recognition, entropy-based anomaly scoring, and graphstructured memory profiling are useful in developing robust behavioural signatures. Deep learning models (CNN, LSTM, RNN, GNN) and machine learning techniques (SVM, DT, RF) are used to improve the accuracy of the classification of polymorphic, metamorphic, and evolving fileless malware. Swarm Optimization Algorithms (SOA) further enhance the detection pipeline by maximizing the features in the detector, and they also adapt to the dynamically varying malware in memory. The experimental findings show better detection, lower analysis latency, and better reconstruction of hidden malicious behaviours in comparison to the conventional forensic techniques.

Fields of Study & Subjects

Advanced Malware Detection Techniques Digital and Cyber Forensics Security and Verification in Computing

Publication details

Journal / source
International Conference on Communication, Computing and Electronics Systems (ICCCES)
Pages
1806–1814
Publication date
21 January 2026
Language
EN