Designing a Robust Machine Learning-Based Framework for Secure Data Transmission in Internet of Things (IoT) Environments - A Multifaceted Approach to Security Challenges

Aug 5, 2026

This research develops a machine learning framework to protect data during transmission in Internet of Things (IoT) environments. The framework targets key security issues by using Random Forest and Support Vector Machine (SVM) to detect abnormal behavior and potential threats in IoT data streams. System effectiveness is assessed with accuracy, precision, recall, and F1-score. Results show Random Forest reaches 93.5% accuracy, slightly higher than SVM at 91.2%, while also detecting cyber-attacks such as DDoS and malware with relatively low false alerts. The study supports integrating machine learning into IoT security to strengthen defenses against emerging attacks and motivate future work using more data for broader use cases.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages10

Reading Time25 min

AccessFree to download

Introduction

Literature Review

Journal of Cyber Security and Risk Auditing Vol.2025, No.4 ISSN: 3079-5354 Journal of Cyber Security and Risk Auditing [https://www.jcsra.thestap.com/](https://www.jcsra.thestap.com/) Designing a Robust Machine Learning-Based Framework for Secure Data Transmission in Internet of Things (IoT) Environments: A Multifaceted Approach to Security Challenges 3 Omar Gheni Abdulateef1 , Atheer Joudah2, Muna Ghazi Abdulsahib2 , Hussein Alrammahi 1 College of Literature, University of Samarra, Salahaldin, Iraq 2College of Computer Science, University of Technology-Iraq, Baghdad, Iraq 3Department of electrical and computer engineering, Altinbaş University, Istanbul, 34000, Turkey ARTICLE INFO Article History Received: 30-06-2025 Revised: 10-08-2025 Accepted: 23-08-2025 Vol.2025, No.4 DOI: [https://doi.org/10.63180/jc](https://doi.org/10.63180/jc)[ ](https://doi.org/10.63180/jc)[sra.thestap.2025.4.6](sra.thestap.2025.4.6) *Corresponding author. Email: omar.ghani@uosamarra.e[du.iq](du.iq) Orcid: [https://orcid.org/0000-](https://orcid.org/0000-)[ ](https://orcid.org/0000-)[0001-7734-6083](0001-7734-6083) This is an open access article under the CC BY 4.0 license ([http://creativecommons.or](http://creativecommons.or)[g/licenses/by/4.0/](g/licenses/by/4.0/) ) . Published by STAP Publisher. ABSTRACT This research develops a machine learning framework for protecting data as it is transmitted in Internet of Things (IoT) configurations. The main objective of the proposed framework to address the major security issues using two intelligent machine learning methods are Random Forest and Support Vector Machine (SVM). They are applied to detect strange behaviour and potential threats within IoT data. The system was evaluated based on accuracy, precision, recall, and F1-score to determine how successful it was. Performance indicated Random Forest performed very well with 93.5% accuracy, slightly higher than SVM 91.2%. The system was also quite good at detecting cyber-attacks such as DDoS and malware, and did not raise many false alerts. This indicates that the system can actually contribute to making IoT much safer, building on what we have in this field. This study implies that incorporating machine learning into IoT security can assist in developing improved defenses against emerging cyber-attacks. In the long term, this research can assist in subsequent studies in order to improve security systems for various uses ofIoT, address existing problems, and utilize more data. Keywords: Internet of Things, Machine Learning, Data Transmission, Security Framework, Random Forest, Anomaly Detection, Cyber Threats. How to cite the article Abdulateef, O. G., Joudah, A. , Abdulsahib, M. G., & Alrammahi, H. (2025) . Designing a Robust Machine Learning-Based Framework for Secure Data Transmission in Internet of Things (IoT) Environments: A Multifaceted Approach to Security Challenges. Journal of Cyber Security and Risk Auditing, 2025(4), 266–275. 1. Introduction The Internet of Things (IoT) has turned into a very significant aspect of today's technology, enabling many devices to communicate with one another and exchange information seamlessly. This technological advancement revolutionized most sectors such as healthcare, smart cities, agriculture, and manufacturing, making processes efficient and convenient for users [1] . As IoT continues to expand, making data secure as it travels becomes increasingly essential. With so many devices holding private information, cyberattacks and data breaches can be a major issue for individuals and businesses. Securing data is really important for ensuring the information in IoT remains accurate, confidential, and always available [2] . Although IoT provides us with amazing things, it has security problems that slow it down. All these problems arise due to weak logins, insecure methods of communicating with one another, and a lack of identical security regulations fo

FAQs

What is the main objective of the proposed framework?

To protect data as it is transmitted in Internet of Things (IoT) environments by addressing major security issues using machine learning methods.

Which machine learning models are used for threat detection?

Random Forest and Support Vector Machine (SVM) are used to detect strange behavior and potential threats in IoT data.

How is the framework’s performance evaluated and what were the results?

Performance is evaluated using accuracy, precision, recall, and F1-score. Random Forest achieved 93.5% accuracy, slightly higher than SVM at 91.2%, with good detection of DDoS and malware and relatively few false alerts.

Aug 5, 2026

Upload by Valentina