Skip to yearly menu bar Skip to main content



Abstract:

This research explores how artificial intelligence (AI) and machine learning (ML) can enhance educational and healthcare outcomes for internally displaced children (IDPs) in Northern Nigeria. The study identifies key barriers to education and healthcare access through quantitative methods, including predictive modelling and data analysis of IDP sites. A mixed-methods approach was employed, integrating machine learning models, stakeholder engagement, and explainable AI techniques to identify key barriers and inform targeted interventions. The dataset, collected from 2,995 internally displaced persons (IDP) sites, encompasses education access, healthcare indicators, and socioeconomic factors. Predictive models were used to analyse school attendance and healthcare utilisation patterns, including Random Forest, Logistic Regression, Support Vector Machine (SVM), XGBoost, and LightGBM. Random Forest demonstrated superior performance in identifying influential features such as household income, teacher availability, and security conditions. SHAP (Shapley Additive Explanations) was applied to ensure model transparency and interpretability. Findings highlight the importance of culturally adaptive AI-driven solutions and gender-sensitive policy recommendations to bridge educational and healthcare gaps in displaced communities. This research underscores the role of ethical AI in addressing inequities and fostering sustainable interventions for vulnerable child populations.

Chat is not available.