Published Articles
Browse published article records from our journals.
Detecting and Quantifying Bias in Machine Learning Models
Mission Franklin (Rivers State University, Rivers State)
Machine learning (ML) systems are increasingly being used to support decisions in areas such as healthcare, finance, education, hiring, and criminal justice. While these technologies can improve efficiency and accuracy, they may also produce unfair outcomes when biases present in data are learned and reproduced by algorithms. Such biases can lead to discrimination against certain individuals or groups, undermining trust in AI-driven systems. This study examines how bias can be detected and measured in machine learning models using established fairness metrics and evaluation techniques. A quantitative approach is adopted to assess model performance across different demographic groups. Metrics such as demographic parity, disparate impact, equal opportunity, and predictive parity are used to evaluate fairness. The findings indicate that machine learning models can exhibit significant disparities among demographic groups even when they achieve high predictive accuracy. The study emphasizes the importance of incorporating fairness assessments throughout the machine learning lifecycle and recommends the adoption of bias mitigation strategies to ensure more equitable outcomes. The research contributes to the development of responsible and trustworthy artificial intelligence systems.
ENSEMBLE MACHINE LEARNING MODELS FOR REAL-TIME NETWORK INTRUSION DETECTION USING SNMP DATA: A SYSTEMATIC REVIEW
Bukola A. Ifedayo-Ojo (Federal University of Technology, Akure, Nigeria.)
Due to the rising sophistication and frequency of cyberattacks targeting business, cloud, Internet of Things (IoT), and software-defined network settings, Network Intrusion Detection Systems (NIDS) remain critical components of current cybersecurity architecture. Traditional intrusion detection systems (IDSs), which rely primarily on single machine learning algorithms or signatures, have limitations in identifying zero-day attacks, reducing false alarms, and adapting to changing network conditions. As a result, ensemble machine learning techniques have attracted significant interest due to their ability to improve detection accuracy and reduce falsepositive rates. However, their effectiveness is strongly dependent on the quality of the input data. Simple Network Management Protocol (SNMP) data offers a lightweight, scalable, and standardised alternative to packet-level traffic analysis in intrusion detection systems. This review employs the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines to identify, evaluate, and synthesise relevant articles published between 2015 and 2025 on the use of ensemble machine learning models for real-time network intrusion detection using SNMP data. This study analyses commonly used ensemble approaches, SNMP Management Information Base (MIB) features, datasets, evaluation metrics and computational efficiency issues. Electronic databases were searched with predefined Boolean searches. Results show that the most widely used methods for SNMP-based intrusion detection include Random Forest, AdaBoost, Gradient Boosting, XGBoost, and stacking ensembles. It also highlights issues such as dataset imbalance, computational overhead, lack of standardised SNMP datasets and real-time deployment limits. The study concludes that combining ensemble learning with SNMP-MIB data is a good approach to developing flexible and scalable intrusion detection systems for modern cybersecurity environments.
Designing Governance Frameworks for Trustworthy and Responsible Artificial Intelligence Systems
Mission Franklin (Rivers State University, Rivers State)
Artificial intelligence is now being widely used in important real-world systems such as healthcare, finance, transportation, and public services. As these systems become more powerful and autonomous, concerns have grown about whether they are always transparent, fair, safe, and accountable in the decisions they make. This paper proposes a structured governance framework designed to support the development and use of trustworthy and responsible AI systems. The framework brings together ethical principles, technical safeguards, and regulatory requirements into a single integrated model. It draws on internationally recognized standards such as the OECD AI Principles and the European Union AI Act to ensure alignment with global best practices. The study presents a multi-layer governance structure that operates across four key dimensions: policy, technical design, system operations, and legal compliance. It also discusses practical challenges that organizations face when implementing AI governance, including bias in algorithms, differences in regulatory requirements across regions, and the difficulty of explaining decisions made by complex AI models. The proposed framework offers a comprehensive approach to ensuring that AI systems are not only effective, but also aligned with human values, ethical expectations, and institutional accountability.