Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: Mar 29, 2026

Author Spotlight: AI-Driven Trypanosome Species Detection from Microscopic Images
08:20

Author Spotlight: AI-Driven Trypanosome Species Detection from Microscopic Images

Published on: October 27, 2023

2.8K

Systematic Evaluation of Machine Learning and Deep Learning Models for IoT Malware Detection Across Ransomware,

Mazdak Maghanaki1, Soraya Keramati2, F Frank Chen1

  • 1Department of Mechanical, Aerospace, and Industrial Engineering, The University of Texas at San Antonio, San Antonio, TX 78249, USA.

Sensors (Basel, Switzerland)
|March 28, 2026
PubMed
Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Evaluating Synthetic Cyber Deception Strategies Under Uncertainty via Game Theory Approach: Linking Information Leakage and Game Outcomes in Cyber Deception.

Sensors (Basel, Switzerland)·2026
See all related articles

Machine learning models, particularly gradient-boosted ensembles like CatBoost and LightGBM, offer superior performance and efficiency for Internet-of-Things (IoT) malware detection compared to deep learning. Optimal model selection depends on specific malware behavior.

Area of Science:

  • Cybersecurity and Machine Learning
  • Internet-of-Things (IoT) Security
  • Malware Detection and Analysis

Background:

  • The proliferation of Internet-of-Things (IoT) devices significantly increases the attack surface for cyber-physical systems.
  • Accurate and efficient malware detection is crucial for securing enterprise and industrial IoT environments.
  • Existing solutions often struggle with the scale and diversity of IoT threats.

Purpose of the Study:

  • To conduct a large-scale, systematic comparison of machine learning (ML) and deep learning (DL) models for IoT malware detection.
  • To evaluate model performance across eight major malware categories: Trojan, Botnet, Ransomware, Rootkit, Worm, Spyware, Keylogger, and Virus.
  • To provide actionable guidance on selecting appropriate models for adaptive IoT malware detection frameworks.
Keywords:
IoT cybersecuritybotnetkeyloggermalware detectionransomwarerootkitspywaretrojanvirusworm

More Related Videos

DNA Virus Detection System Based on RPA-CRISPR/Cas12a-SPM and Deep Learning
04:17

DNA Virus Detection System Based on RPA-CRISPR/Cas12a-SPM and Deep Learning

Published on: May 10, 2024

1.7K

Related Experiment Videos

Last Updated: Mar 29, 2026

Author Spotlight: AI-Driven Trypanosome Species Detection from Microscopic Images
08:20

Author Spotlight: AI-Driven Trypanosome Species Detection from Microscopic Images

Published on: October 27, 2023

2.8K
DNA Virus Detection System Based on RPA-CRISPR/Cas12a-SPM and Deep Learning
04:17

DNA Virus Detection System Based on RPA-CRISPR/Cas12a-SPM and Deep Learning

Published on: May 10, 2024

1.7K

Main Methods:

  • A dataset of 50,000 executable samples (8,000 malware, 42,000 benign) was created from the Any.Run platform.
  • Static and behavioral telemetry were extracted via sandbox execution, followed by feature selection yielding 47 diagnostic features.
  • 27 ML and 18 DL models were trained and evaluated on the curated feature set.

Main Results:

  • Ensemble tree-based ML models, specifically CatBoost and LightGBM, demonstrated superior accuracy and low false-positive rates.
  • Optimal model choice varied by malware type: CatBoost for Trojan/Spyware, LightGBM for Botnet, XGBoost for Worm, Extra Trees for Rootkit, Random Forest for Keylogger.
  • Deep learning models showed competitive performance only for specific categories (TabNet for Ransomware, FT-Transformer for Virus), with ML models generally offering better computational efficiency.

Conclusions:

  • Gradient-boosted ensemble ML models are recommended as the primary choice for IoT malware detection due to their performance and efficiency.
  • Deep learning models may be selectively deployed when category-specific performance gains outweigh the increased computational cost.
  • The study provides a data-driven framework for selecting effective IoT malware detection strategies.