Adaptive Machine Learning for Real-Time Intrusion Detection in IoT

Ouku Bhulakshmi, Muddam Anusha, Ramisetty Somesh, Surasetty Badrinath, Mangali Madhan Gopal, Pattan Thoufiq Khan

2025

Abstract

The Internet of Things (IoT) gadgets are extensively used throughout several domains, providing numerous conveniences to individuals' lives. However, the extensive deployment of IoT devices has made maintaining these systems against cyber-attacks a primary concern for researchers. IoT devices possess limited computing capabilities and storage resources, leading to inadequate security defense mechanisms and heightened vulnerability to malware and device assaults. Current IoT-focused intrusion detection solutions often merely identify specific malicious attempts or require intricate models and substantial processing resources to achieve elevated detection accuracy. In this study, we utilize three datasets: BoT-IoT, MedBIoT, and MQTT-IoT-IDS 2020. We implement various algorithms, including Decision Tree, Random Forest, KNN, XGBoost, DNN, CNN, and advanced ensemble techniques such as Stacking Classifier (DT + RF with LightGBM) and CNN + LSTM. Our results demonstrate that the Stacking Classifier achieved the highest performance, with superior accuracy, precision, recall, and F1 score. The Stacking Classifier achieved a high accuracy of 100% in BoT-IoT and MedBIoT, and 92.3% in MQTT-IoT-IDS 2020, effectively enhancing the robustness and accuracy of IoT intrusion detection in resource-constrained environments. This method provides a lightweight and efficient solution to improve security measures for IoT devices.

Paper Citation

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25
TI - Adaptive Machine Learning for Real-Time Intrusion Detection in IoT
SN - 978-989-758-777-1
AU - Bhulakshmi O.
AU - Anusha M.
AU - Somesh R.
AU - Badrinath S.
AU - Gopal M.
AU - Khan P.
PY - 2025
SP - 654
EP - 662
DO - 10.5220/0013887900004919
PB - SciTePress

in Harvard Style

Bhulakshmi O., Anusha M., Somesh R., Badrinath S., Gopal M. and Khan P. (2025). Adaptive Machine Learning for Real-Time Intrusion Detection in IoT. In Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25; ISBN 978-989-758-777-1, SciTePress, pages 654-662. DOI: 10.5220/0013887900004919

in Bibtex Style

@conference{icrdicct`2525,
author={Ouku Bhulakshmi and Muddam Anusha and Ramisetty Somesh and Surasetty Badrinath and Mangali Madhan Gopal and Pattan Thoufiq Khan},
title={Adaptive Machine Learning for Real-Time Intrusion Detection in IoT},
booktitle={Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25},
year={2025},
pages={654-662},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013887900004919},
isbn={978-989-758-777-1},
}