Performance Analysis of Selected Machine Learning Algorithms in the prediction of Man-in-the-Middle in Internet of Things Environment

Aug 5, 2026

Upload by Ezra

Concerns over the rapid expansion and vulnerability of Internet of Things (IoT) devices to cyberattacks motivate this performance study on Man-in-the-Middle (MitM) threats. Logistic Regression, Decision Trees, and K-Nearest Neighbors are evaluated and compared using accuracy, precision, recall, F1-score, and error rate on a Kaggle dataset containing both normal and attacked samples. Decision Trees achieve the highest MitM prediction accuracy (99.42%) with strong balance across precision, F1-score, and recall, and the lowest error rate (0.0058). Findings support selecting appropriate machine-learning models to strengthen IoT security systems and indicate future work on transfer other techniques for MitM detection.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages19

Reading Time48 min

AccessFree to download

Introduction

Internet of Things (IoT) fundamentals

Man-in-the-Middle (MiTM) attack overview

Intrusion Detection Systems (IDS) and limitations for IoT

Machine learning approaches for MitM detection

Performance Analysis of Selected Algorithms

Covenant Journal of Informatics & Communication Technology. Vol. 12, No. 2, December, 2024 ISSN: print 2354 – 3566 electronics 2354 – 3507 DOI: An Open Access Journal Available Online Performance Analysis of selected Machine Learning Algorithms in the prediction of Man in the Middle in Internet of Things Environment Stephen A. Mogaji, Olaiya Folorunsho, Yetunde Daramola, Timothy T. Odufuwa Department of Computer Sciences, Federal University, Oye Ekiti, Nigeria. [stephen.mogaji@fuoye.edu.ng](stephen.mogaji@fuoye.edu.ng), [olaiya.folorunsho@fuoye.edu.ng](olaiya.folorunsho@fuoye.edu.ng), [comfort.daramola@fuoye.edu.ng](comfort.daramola@fuoye.edu.ng), [tolulope.odufuwa@fuoye.edu.ng](tolulope.odufuwa@fuoye.edu.ng) [Received: xx.xx.xxxx Accepted: xx.xx.xxxx](Received: xx.xx.xxxx Accepted: xx.xx.xxxx) Publication: December 2024 Abstract— Concerns have been expressed over Internet of Things (IoT) devices'growing prevalence and susceptibility to cyberattacks, namely Man-in-the-Middle (MitM) assaults. The performance of selected machine learning algorithms: Logistic Regression, Decision Trees, and K-Nearest Neighbors were analyzed and compared using accuracy, precision, recall, F1-score, and error rate using a dataset comprising normal and attacked data sets from Kaggle. According to the research findings, the Decision Tree algorithm outperformed other selected algorithms in terms of MitM attack prediction accuracy of 99.42% and a good balance between precision, F1-score, and recall, with the lowest error rate of 0. 0058. The results of the study improve the security and reliability of IoT applications by aiding in the creation of efficient MitM attack prediction systems for IoT environments. The findings also emphasize how crucial it is to choose the best machine-learning algorithm for a given IoT security task. Investigating the use of transfer other techniques in MitM attack detection for IoT contexts is one area of future research. Keywords/Index Terms— Prediction, MiTM, Decision Tree, Logistic regression, KNN 1. Introduction The Internet of Things (IoT) permits seamless communications, from smart household appliances to industrial machines enabling them to communicate and interact with each other over the internet. According to (Alexander, 2024), the Internet of Things (IoT) is a system of intelligent things that includes sensors, actuators, programmable central microcontrollers, and other processors that may be wirelessly connected to routers and gateways. The core of IoT networks are sensors, which are in charge of monitoring and collecting data on changes in the physical environment, such as temperature, motion, humidity, and pressure, and convert the detected changes into a format that can be understood and processed by the microcontroller or processor, transmitting the data in either digital or analog signals, enabling the IoT network to make decisions and operate efficiently. The microcontrollers or processors utilized in both embedded systems and various applications and IoT networks are designed to be efficient and are often battery-powered, which also form the core of the network, are typically batterypowered and resource-constrained in terms of power consumption, Random Access Memory (RAM), and Read-Only Memory (ROM) . These resource constraints require efficient processing, memory management, and power optimization to ensure reliable performance and extended battery life in IoT devices (Aeris, 2024) . This network is made up of various connected "things"including devices like networked household devices, portable technology, and networked vehicles, all of which communicate with each other and with centralized systems to automate and streamline processes. When a felonious individual places himself in the center of a discussion between a user and an application, either to eavesdrop or to pretend to be one of the parties, creating the impression that a legitimate information exchange is taking

FAQs

Which machine learning algorithms are analyzed for MitM prediction in IoT environments?

The study compares Logistic Regression, Decision Trees, and K-Nearest Neighbors using evaluation metrics including accuracy, precision, recall, F1-score, and error rate.

How is the dataset for the MitM prediction experiments constructed?

The models are tested using a Kaggle dataset that includes both normal data and MitM attacked data.

What algorithm performs best for MitM attack prediction and what are its key results?

Decision Trees outperform the other algorithms, reaching 99.42% accuracy, a strong balance across precision/F1-score/recall, and the lowest error rate of 0.0058.

Aug 5, 2026

Upload by Ezra