Application of deep learning in malware detection: a review

Abstract The defense of malware remains an important research hotspot in the field of cyberspace security. Recognizing its profound research significance, our defense against malware is still an important research hotspot in the field of cyberspace security. According to several recent surveys, glob...

Full description

Saved in:
Bibliographic Details
Main Authors: Yafei Song, Dandan Zhang, Jian Wang, Yanan Wang, Yang Wang, Peng Ding
Format: Article
Language:English
Published: SpringerOpen 2025-04-01
Series:Journal of Big Data
Subjects:
Online Access:https://doi.org/10.1186/s40537-025-01157-y
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849713674819731456
author Yafei Song
Dandan Zhang
Jian Wang
Yanan Wang
Yang Wang
Peng Ding
author_facet Yafei Song
Dandan Zhang
Jian Wang
Yanan Wang
Yang Wang
Peng Ding
author_sort Yafei Song
collection DOAJ
description Abstract The defense of malware remains an important research hotspot in the field of cyberspace security. Recognizing its profound research significance, our defense against malware is still an important research hotspot in the field of cyberspace security. According to several recent surveys, global infrastructure is increasingly attacked by cyber crimes, and the damage of various malicious attacks to countries and even individuals cannot be underestimated, even on the rise. There is an urgent need to adopt advanced tools for early detection of malware and its variants to help researchers take early steps to defend against it. Its broad approach will help the early malware to detect and identify the behavioral patterns of large amounts of malicious data, and the discipline of artificial intelligence offers broad research potential. The results of these tests will help researchers make decisions and early detection, effectively defense against malware. This work compares and reports a classification of malware detection work based on deep learning algorithms. The 2011–2025 articles were considered, and the latest work focused on the literature for the 2018–2025 years; after screening, 72 articles were selected for the initial study. Future researchers will benefit from this review by better understanding current deep learning models in the field of malware detection. The review includes common methods such as convolutional neural networks, recurrent neural networks and generative adversarial networks, focusing on feature extraction techniques such as sequence features, image visualization and data enhancement. The survey summarizes the metrics used to report the accuracy. In addition, it highlights prominent publishers, journals and conferences as platforms for the evaluation of academic works. Taken together, this will help researchers at the current stage gain insight into the unresolved challenges or barriers faced by previous researchers. Among these, the most common problem is the lack of broader and consistent datasets, followed by the need for existing models for further improvement.
format Article
id doaj-art-3e1b5bdffe82437caec87a0639a29905
institution DOAJ
issn 2196-1115
language English
publishDate 2025-04-01
publisher SpringerOpen
record_format Article
series Journal of Big Data
spelling doaj-art-3e1b5bdffe82437caec87a0639a299052025-08-20T03:13:54ZengSpringerOpenJournal of Big Data2196-11152025-04-0112112910.1186/s40537-025-01157-yApplication of deep learning in malware detection: a reviewYafei Song0Dandan Zhang1Jian Wang2Yanan Wang3Yang Wang4Peng Ding5Institute of Air Defense and Anti-Missile, Air Force Engineering UniversityInstitute of Air Defense and Anti-Missile, Air Force Engineering UniversityInstitute of Air Defense and Anti-Missile, Air Force Engineering UniversityInstitute of Air Defense and Anti-Missile, Air Force Engineering UniversityInstitute of Air Defense and Anti-Missile, Air Force Engineering UniversityInstitute of Air Defense and Anti-Missile, Air Force Engineering UniversityAbstract The defense of malware remains an important research hotspot in the field of cyberspace security. Recognizing its profound research significance, our defense against malware is still an important research hotspot in the field of cyberspace security. According to several recent surveys, global infrastructure is increasingly attacked by cyber crimes, and the damage of various malicious attacks to countries and even individuals cannot be underestimated, even on the rise. There is an urgent need to adopt advanced tools for early detection of malware and its variants to help researchers take early steps to defend against it. Its broad approach will help the early malware to detect and identify the behavioral patterns of large amounts of malicious data, and the discipline of artificial intelligence offers broad research potential. The results of these tests will help researchers make decisions and early detection, effectively defense against malware. This work compares and reports a classification of malware detection work based on deep learning algorithms. The 2011–2025 articles were considered, and the latest work focused on the literature for the 2018–2025 years; after screening, 72 articles were selected for the initial study. Future researchers will benefit from this review by better understanding current deep learning models in the field of malware detection. The review includes common methods such as convolutional neural networks, recurrent neural networks and generative adversarial networks, focusing on feature extraction techniques such as sequence features, image visualization and data enhancement. The survey summarizes the metrics used to report the accuracy. In addition, it highlights prominent publishers, journals and conferences as platforms for the evaluation of academic works. Taken together, this will help researchers at the current stage gain insight into the unresolved challenges or barriers faced by previous researchers. Among these, the most common problem is the lack of broader and consistent datasets, followed by the need for existing models for further improvement.https://doi.org/10.1186/s40537-025-01157-yMalwareMalware classificationMalware detectionDeep learningReview
spellingShingle Yafei Song
Dandan Zhang
Jian Wang
Yanan Wang
Yang Wang
Peng Ding
Application of deep learning in malware detection: a review
Journal of Big Data
Malware
Malware classification
Malware detection
Deep learning
Review
title Application of deep learning in malware detection: a review
title_full Application of deep learning in malware detection: a review
title_fullStr Application of deep learning in malware detection: a review
title_full_unstemmed Application of deep learning in malware detection: a review
title_short Application of deep learning in malware detection: a review
title_sort application of deep learning in malware detection a review
topic Malware
Malware classification
Malware detection
Deep learning
Review
url https://doi.org/10.1186/s40537-025-01157-y
work_keys_str_mv AT yafeisong applicationofdeeplearninginmalwaredetectionareview
AT dandanzhang applicationofdeeplearninginmalwaredetectionareview
AT jianwang applicationofdeeplearninginmalwaredetectionareview
AT yananwang applicationofdeeplearninginmalwaredetectionareview
AT yangwang applicationofdeeplearninginmalwaredetectionareview
AT pengding applicationofdeeplearninginmalwaredetectionareview