Analyzing the Techniques & Algorithms for Malware Detection Using Deep Learning Neural Networks

DOI

10.2991/978-94-6239-610-4_42How to use a DOI?

Keywords

Malware; Hybrid Analysis; CNN; LSTM; DNN

Abstract

Cybersecurity faces an ongoing and quickly changing challenge from malicious software (malware). Usually, traditional methods of detection rely on static or dynamic analysis methods, neither of which provide complete protection against all versions of malware. With static analysis, malware characteristics like API calls, permissions, and opcodes are extracted without actually executing the malware file; however, these types of analyses are easily obfuscated and packed. While static malware detection methods rely on analyzing variables within a sample without executing it, dynamic malware detection methods utilize the execution of a sample to monitor what happens at runtime (e.g., when a malware sample makes system calls and connects to remote servers). Because of the greater resiliency of dynamic analysis methods over static analyses, they require a significantly higher amount of resources and therefore have created techniques for creating “sandbox” environments where malware can execute. In addition, as with all detection methods, attackers can create unique ways to defeat detection via evasive behaviour and anti-reverse engineering methods. To solve these problems, we created a hybrid malware detection framework that utilizes both types of analysis in combination with deep-learning approaches. This framework utilizes CNNs, LSTMs, and DNNs to capture spatial relationships, temporal relationships, and high- level representations of both types of features within the analysed Windows API dataset(s). Using Evaluation metrics we show that the hybrid detection framework is capable of reliably classifying malware samples as either benevolent or malicious. By providing a hybrid system for the scalable, rapid detection of modern malware threats in today’s increasingly complex computer environments, our framework serves as an opportunity to improve malware detection systems.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Arun Singh Chouhan
AU  - Bollamalli Pravalika
PY  - 2026
DA  - 2026/05/05
TI  - Analyzing the Techniques & Algorithms for Malware Detection Using Deep Learning Neural Networks
BT  - Proceedings of the First International Conference on Advances in Forensics and Cyber Technologies (ICFACT 2025)
PB  - Atlantis Press
SP  - 488
EP  - 497
SN  - 2352-538X
UR  - https://doi.org/10.2991/978-94-6239-610-4_42
DO  - 10.2991/978-94-6239-610-4_42
ID  - Chouhan2026
ER  -