A Comparative Study of Anomaly Detection Techniques for IoT Security Using Adaptive Machine Learning for IoT Threats

Anomaly detection is a critical aspect of various applications, including security, healthcare, and network monitoring. In this study, we introduce FusionNet, an innovative ensemble model that combines the strengths of multiple machine learning algorithms, namely Random Forest, K-Nearest Neighbors,...

Full description

Saved in:
Bibliographic Details
Main Author: Dheyaaldin Alsalman
Format: Article
Language:English
Published: IEEE 2024-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10415174/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1832583974756548608
author Dheyaaldin Alsalman
author_facet Dheyaaldin Alsalman
author_sort Dheyaaldin Alsalman
collection DOAJ
description Anomaly detection is a critical aspect of various applications, including security, healthcare, and network monitoring. In this study, we introduce FusionNet, an innovative ensemble model that combines the strengths of multiple machine learning algorithms, namely Random Forest, K-Nearest Neighbors, Support Vector Machine, and Multi-Layer Perceptron, for enhanced anomaly detection. FusionNet’s architecture leverages the diversity of these algorithms to achieve high accuracy and precision. We evaluate FusionNet’s performance on two distinct datasets, Dataset 1 and Dataset 2, and compare it with traditional machine learning models, including SVM, KNN, and RF. The results demonstrate that FusionNet consistently outperforms these models across both datasets in terms of accuracy, precision, recall, and F1 score. On Dataset 1, FusionNet achieves an accuracy of 98.5% and on Dataset 2, it attains an accuracy of 99.5%. FusionNet’s remarkable ability to detect anomalies with exceptional accuracy underscores its potential for real-world applications. This study highlights the significance of FusionNet as a robust model for anomaly detection and provides insights into its superior performance over traditional models. The results emphasize the promising prospects of FusionNet in security, healthcare, and other domains where accurate anomaly detection is crucial.
format Article
id doaj-art-52cf265d8e9c4279bb5505e4914600c3
institution Kabale University
issn 2169-3536
language English
publishDate 2024-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-52cf265d8e9c4279bb5505e4914600c32025-01-28T00:00:48ZengIEEEIEEE Access2169-35362024-01-0112147191473010.1109/ACCESS.2024.335903310415174A Comparative Study of Anomaly Detection Techniques for IoT Security Using Adaptive Machine Learning for IoT ThreatsDheyaaldin Alsalman0https://orcid.org/0000-0002-8493-2758Department of Cybersecurity, School of Engineering, Computing and Design, Dar Al-Hekma University, Jeddah, Saudi ArabiaAnomaly detection is a critical aspect of various applications, including security, healthcare, and network monitoring. In this study, we introduce FusionNet, an innovative ensemble model that combines the strengths of multiple machine learning algorithms, namely Random Forest, K-Nearest Neighbors, Support Vector Machine, and Multi-Layer Perceptron, for enhanced anomaly detection. FusionNet’s architecture leverages the diversity of these algorithms to achieve high accuracy and precision. We evaluate FusionNet’s performance on two distinct datasets, Dataset 1 and Dataset 2, and compare it with traditional machine learning models, including SVM, KNN, and RF. The results demonstrate that FusionNet consistently outperforms these models across both datasets in terms of accuracy, precision, recall, and F1 score. On Dataset 1, FusionNet achieves an accuracy of 98.5% and on Dataset 2, it attains an accuracy of 99.5%. FusionNet’s remarkable ability to detect anomalies with exceptional accuracy underscores its potential for real-world applications. This study highlights the significance of FusionNet as a robust model for anomaly detection and provides insights into its superior performance over traditional models. The results emphasize the promising prospects of FusionNet in security, healthcare, and other domains where accurate anomaly detection is crucial.https://ieeexplore.ieee.org/document/10415174/CybersecurityIoMTintrusion detectionmachine learningadaptive learningSVM
spellingShingle Dheyaaldin Alsalman
A Comparative Study of Anomaly Detection Techniques for IoT Security Using Adaptive Machine Learning for IoT Threats
IEEE Access
Cybersecurity
IoMT
intrusion detection
machine learning
adaptive learning
SVM
title A Comparative Study of Anomaly Detection Techniques for IoT Security Using Adaptive Machine Learning for IoT Threats
title_full A Comparative Study of Anomaly Detection Techniques for IoT Security Using Adaptive Machine Learning for IoT Threats
title_fullStr A Comparative Study of Anomaly Detection Techniques for IoT Security Using Adaptive Machine Learning for IoT Threats
title_full_unstemmed A Comparative Study of Anomaly Detection Techniques for IoT Security Using Adaptive Machine Learning for IoT Threats
title_short A Comparative Study of Anomaly Detection Techniques for IoT Security Using Adaptive Machine Learning for IoT Threats
title_sort comparative study of anomaly detection techniques for iot security using adaptive machine learning for iot threats
topic Cybersecurity
IoMT
intrusion detection
machine learning
adaptive learning
SVM
url https://ieeexplore.ieee.org/document/10415174/
work_keys_str_mv AT dheyaaldinalsalman acomparativestudyofanomalydetectiontechniquesforiotsecurityusingadaptivemachinelearningforiotthreats
AT dheyaaldinalsalman comparativestudyofanomalydetectiontechniquesforiotsecurityusingadaptivemachinelearningforiotthreats