M. A. Ayanwale

Review Open access Sep 2026

Digital Governance as Institutional Innovation for Sustainable Public-Sector Transformation in Africa: A PRISMA-Guided Systematic Literature Review and Framework for Inclusive and Decentralized Governance

Digital governance has increasingly emerged as a critical institutional mechanism for strengthening public administration, enhancing accountability, improving service delivery, and advancing inclusive development across Africa. Governments across the continent have expanded investments in e-governance systems, digital public infrastructure, artificial intelligence (AI)-enabled administrative systems, interoperable service platforms, and data-driven governance reforms as part of broader modernization and Sustainable Development Goal (SDG) agendas. Despite these developments, digital governance outcomes remain uneven and are frequently constrained by fragmented governance architectures, weak institutional coordination, administrative capacity deficits, regulatory limitations, and persistent socio-economic inequalities. To address these challenges, this study conducts a qualitative Systematic Literature Review (SLR) guided by PRISMA 2020 reporting standards to examine how digital technologies interact with governance systems, institutional structures, administrative capability, and socio-technical inequalities across African public sectors. The review synthesized 42 included studies published between 2015 and 2025 using structured Boolean search strategies, predefined inclusion and exclusion criteria, abductive thematic synthesis, and framework-oriented analytical procedures. The findings identify three interconnected governance constraints shaping digital transformation outcomes across African public sectors: fragmented and centralized governance arrangements; institutional and administrative capability deficits; and persistent digital inequalities and exclusionary governance systems. Although digital governance reforms demonstrate important potential for improving transparency, interoperability, citizen participation, financial inclusion, and administrative efficiency, sustainable transformation outcomes depend heavily on institutional coordination, adaptive governance systems, digital inclusion, regulatory effectiveness, and accountable AI governance arrangements. Building on the synthesized evidence, the study develops a decentralized AI-enabled digital public governance framework linking decentralization, interoperability, institutional coordination, digital inclusion, adaptive governance, and ethical AI governance to sustainable public-sector transformation outcomes. The study contributes theoretically by reconceptualizing digital governance as an institutionally embedded governance transformation process rather than merely a technological modernization agenda. The findings further contribute to sustainability debates by demonstrating how inclusive and decentralized digital governance systems can strengthen institutional resilience, public-sector innovation, and sustainable development outcomes across diverse African governance contexts.

V. Egba, M. A. Ayanwale, I. Ukeje et al. · Sustainability · 0 citations

Open access Jul 2026

Fairness-Aware Machine Learning for Ethical Student Performance Prediction in Education

The integration of artificial intelligence (AI) and machine learning (ML) into education has transformed how student performance is predicted and monitored. Despite these advances, concerns regarding fairness, transparency, interpretability, and potential demographic bias remain significant challenges in educational prediction systems. This study developed ethically aligned and interpretable ML models for predicting student academic performance using only behavioural engagement and academic context variables. The open-access xAPI-Edu-Data dataset containing 480 student records was obtained from Kaggle. Four supervised algorithms, Multinomial Logistic Regression, Decision Tree, Random Forest, and XGBoost, were implemented in Python 3.11 using Scikit-learn, SHAP, and LIME frameworks. To minimise data leakage, all preprocessing operations, including standardisation and categorical encoding, were embedded within a Scikit-learn Pipeline fitted exclusively on the training folds. A stratified train/validation/test split (70%/15%/15%) with random_state = 42 was employed, while hyperparameters were optimised using five-fold cross-validation on the training and validation sets only. Model performance was evaluated using accuracy, macro F1-score, ROC-AUC, Brier Score, and Expected Calibration Error (ECE). Results showed that Random Forest achieved the best overall performance and calibration, while Logistic Regression provided superior interpretability. Across all models, visited_resources and raisedhands consistently emerged as the strongest predictors of academic achievement, emphasising the importance of student engagement behaviours. The study demonstrates that fairness-by-design prediction systems that exclude demographic variables can support equitable and actionable early-warning interventions. The novelty of the study lies in its ethically grounded and deployment-oriented framework for interpretable educational prediction in resource-constrained contexts.

M. A. Ayanwale, I. J. Chikezie · Journal of Computer Adaptive... · 0 citations