Claim Missing Document
Check
Articles

Found 5 Documents
Search

Machine Learning Algorithms for Anomaly Detection in IoT Networks – A Review Muhammad Mamman Kontagora; Bartholomew Idoko
African Multidisciplinary Journal of Sciences and Artificial Intelligence Vol 1 No 2 (2024): African Multidisciplinary Journal of Sciences and Artificial Intelligence
Publisher : Darul Yasin Al Sys

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.58578/amjsai.v1i2.4014

Abstract

Internet of Things (IoT) wide applications has significantly increased the need for robust anomaly detection to safeguard against countless security breaches. This paper presents a review that examines the effectiveness of hybrid solutions incorporating supervised and unsupervised machine learning models for enhancing IoT security. The review consolidates insights from a range of studies employing models such as Random Forest (RF), Support Vector Machine (SVM), k-nearest Neighbors (k-NN), and Gaussian Mixture Models (GMM). It integrates the findings of diverse research, emphasizing improvements in terms of detection accuracy and computational demands. The study delineates challenges in the field to evaluate the efficacy of hybrid techniques and their potential for immediate IoT security applications. Moreover, future research directions encompass the exploration of new algorithms and the integration of these approaches within dynamic IoT data streams.
Detection of Malware Attacks in Medical Mechatronics Distribution System Using Support Vector Machine Bartholomew Idoko; Okoro Denis Isah; Sampson Agada; Samuel Owoicho Olofu
African Multidisciplinary Journal of Sciences and Artificial Intelligence Vol 2 No 3 (2025): African Multidisciplinary Journal of Sciences and Artificial Intelligence
Publisher : Darul Yasin Al Sys

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.58578/amjsai.v2i3.7231

Abstract

This study examines the cybersecurity challenges and solutions associated with medical mechatronics devices, which increasingly operate autonomously through advances in robotics, artificial intelligence (AI), and machine-to-machine communication. As the number of interconnected devices grows—from smart grids and home appliances to medical equipment and sensor–actuator testers—ensuring secure and trustworthy communication becomes critical. A sustainable defensive strategy for medical mechatronics requires robust systems capable of malware analysis and detection, informed by an understanding of cyber-attack stages such as reconnaissance, weaponization, delivery, exploitation, installation, and command and control. Traditional malware detection systems struggle with obfuscated malware, making AI and machine learning (ML) more effective tools for accurate detection and classification. This research proposes the use of a Support Vector Machine (SVM) model with a novel metric to enhance malware detection in medical mechatronics devices, thereby strengthening confidentiality, integrity, availability, and digital trust. The proposed SVM-based approach was compared with established SVM algorithms using a real dataset from medical mechatronics distribution systems across federal medical centers in Nigeria. Findings demonstrate the potential of the model to improve malware detection accuracy and compliance with digital sovereignty standards, offering practical insights for enhancing cybersecurity in critical healthcare technologies.
Cryptographic System for Mobile Application (Automated Resume Builder) Bartholomew Idoko; Sampson Agada; Okoro Denis Isah; Chika Patricia Bossah
Kwaghe International Journal of Sciences and Technology Vol 2 No 3 (2025): Kwaghe International Journal of Sciences and Technology
Publisher : Darul Yasin Al Sys

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.58578/kijst.v2i3.7228

Abstract

This paper presents a secure and user-friendly approach for developing an automated mobile application system, using a resume builder as a case study. The proposed system automates the construction of resumes by utilizing applicants’ information as input, allowing users to create, edit, delete, read, and save resumes in PDF format, supported by login and signup via OTP verification. To enhance security, the study introduces a two-factor authentication (TFA) scheme that integrates a cryptographic-compatible device and a password, offering stronger protection against risks such as communication breaches, device or server vulnerabilities, and offline or online credential attacks. The TFA is implemented through shared access signature (SAS) message authentication or other PIN-based authentication methods. The system architecture incorporates an enhanced cryptographic framework adaptable to various password-based client–server authentication protocols, reducing reliance on less secure single-layer password systems. Data encryption is handled using the Advanced Encryption Standard (AES), chosen over 3DES for its superior processing efficiency, while the Message-Digest Method (MD5) algorithm is used to hash user-defined encryption keywords. All server-side data, including encryption keys, remain encrypted, ensuring that unauthorized access yields no advantage. By enabling users to encrypt and decrypt data with AES and securing encryption keys via MD5 hashing, the system improves both privacy and security in mobile applications. The study contributes to secure software design by demonstrating how cryptographic methods can be modularly integrated into mobile systems, addressing the cybersecurity gaps of conventional job search and resume platforms.
Development of a Framework for Cybersecurity Risk Assessment in the Maritime Industry Using Machine Learning Techniques Bartholomew Idoko; Kenneth Nwankwo
Kwaghe International Journal of Engineering and Information Technology Vol 2 No 2 (2025): Kwaghe International Journal of Engineering and Information Technology
Publisher : Darul Yasin Al Sys

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.58578/kijeit.v2i2.5342

Abstract

This study assesses the level of cybersecurity risk inherent in the maritime industry in order to improve process in the sector. The maritime sector has continued to witness cyber incidents due to its importance to national economy. Also, the growing dependence of the sector on information and communications technology (ICT), as a result of increased automation, has greatly exacerbated the threats. The underlying cyber infrastructure with its expanding threat landscape and vulnerabilities have also further exacerbated the risk landscape in the sector. More so, the dearth of empirical studies in this domain is an indication of knowledge gap occasioned by non-availability of empirical data on how organizations in this sector manage cybersecurity risk. That is, how organizational operations and technological assets, individuals and processes affect the sector. Thus, the study has identified and established the cybersecurity risks specific to the maritime sector and gauged the gap based on people, process and technology elements of cybersecurity. This study uses Artificial Intelligence, machine learning model in particular to carry out the assessment. The study identified how organizations applied security controls in the sector using the metrics of people, process and technology. The risk was analysed and graded into very high, high, moderate, low and very low from the established risk factors like threat and vulnerabilities. We used k-nearest neigbour and factorization methods for model training and risk ratings. The findings showed that the maritime sector has a high cybersecurity risk rating. This knowledge and the recommendations that followed, will help deepen the understanding of cybersecurity risks in the maritime sector as well as improve maritime process, its potential effects on service delivery, national security and economic wellbeing of the nation.
Swarm Intelligence-Based Intrusion Detection Framework Using Neural Network & Bee Colony Optimiation Kenneth Nwankwo; Bartholomew Idoko
Kwaghe International Journal of Engineering and Information Technology Vol 2 No 2 (2025): Kwaghe International Journal of Engineering and Information Technology
Publisher : Darul Yasin Al Sys

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.58578/kijeit.v2i2.5452

Abstract

An Intrusion Detection System (IDS) serves as a critical defense mechanism for safeguarding networks against unauthorized activities and cyber attacks. However, the processing of sophisticated datasets with contemporary detection methodologies often presents challenges due to their intricate scale, complicating the identification of complex threats. This study aims to enhance IDS operational efficacy through the development of a novel method integrating Bee Colony Optimization (BCO) and Neural Networks (NN). Employing a quasi-experimental design, the research evaluates the system's performance, demonstrating that the integration of BCO significantly optimizes neural network functionality, thereby improving both the speed of attack detection and the accuracy of feature selection. Utilizing the NSL-KDD dataset, the proposed framework notably minimizes false alerts while augmenting overall detection accuracy levels. The findings underscore that advancements in cybersecurity systems can be achieved through the synergy of Neural Networks and Swarm Intelligence technology, providing effective solutions for real-time intrusion detection systems. This research not only contributes to the theoretical understanding of IDS optimization but also has practical implications for enhancing cybersecurity measures in various organizational contexts.