Sign in Subscribe
Machine learning

Advances in Machine Learning-Based Intrusion Detection in IoT Networks Revealed

Strategic Insights

— 1 min read

Research from the Technical University of Denmark (DTU) has uncovered trends and challenges in machine learning-based intrusion detection in Internet of Things (IoT) networks. The study, which analyzed datasets and machine learning techniques used from 2019 to 2024, found that certain datasets, such as BoT-IoT and TON_IoT, combined with Decision Tree (DT) and Random Forest (RF) models, achieved high median accuracy rates of 99%. The research highlights the need for robust and scalable deployment options, with Software-Defined Networks (SDNs) offering flexibility, edge computing being extensively explored in cloud environments, and blockchain-integrated networks emerging as a promising approach for enhancing security.

Key Takeaways:

  • The study analyzed 33 papers on machine learning-based DDoS attacks in IoT networks, covering deployment options, datasets, and machine learning techniques used between 2019 and 2024.
  • The top-performing datasets were BoT-IoT and TON_IoT, achieving high median accuracy rates of 99% when combined with Decision Tree (DT) and Random Forest (RF) models.
  • The research found that hardware limitations led to a preference for lightweight machine learning solutions and preprocessed datasets.
  • Current trends indicate that larger or industry-specific datasets will continue to gain popularity alongside more complex machine learning models, such as deep learning.
  • The study highlighted the importance of robust and scalable deployment options, with Software-Defined Networks (SDNs) offering flexibility, edge computing being extensively explored in cloud environments, and blockchain-integrated networks emerging as a promising approach for enhancing security.
  • The research concluded that there is a need for more robust and scalable deployment options to enhance security in IoT networks.

Statistics:

  • 33 papers were analyzed in the study on machine learning-based DDoS attacks in IoT networks.
  • 99% median accuracy rate was achieved by using BoT-IoT and TON_IoT datasets combined with Decision Tree (DT) and Random Forest (RF) models.
  • 2019-2024 was the time period analyzed in the study.
  • 4 datasets (BoT-IoT and TON_IoT) were found to be the top-performing datasets in the study.
  • 2 machine learning models (Decision Tree and Random Forest) were used in the study to achieve high median accuracy rates.

Sources:

  • Advancements in Machine Learning-Based Intrusion Detection in IoT: Research Trends and Challenges. Algorithms, 2025,18(4):209.
  • MDPI AG.
  • Algorithms - http://www.mdpi.com/journal/algorithms.
  • DOI: 10.3390/a18040209.

Read more