SURVEY OF DIGITAL TECHNOLOGIES ADOPTION PREDICTION IN GIDAN MADI USING MACHINE LEARNING MODELS

Abstract

Digital technologies are widely recognized as catalysts for economic growth, educational access, and social inclusion; however, their adoption remains persistently low in many rural communities in developing regions. In areas such as Gidan Madi, Sokoto State, Nigeria, limited digital literacy, poor infrastructure, and socio-economic constraints continue to widen the digital divide, restricting access to essential digital services. This study aims to develop a machine learning–based predictive framework to identify the key determinants of digital tool adoption in underserved, low-literacy rural communities and to support evidence-based digital inclusion initiatives. A household survey dataset comprising 1,450 records was analyzed using four supervised machine learning algorithms: Logistic Regression, Random Forest, k-Nearest Neighbors, and Gradient Boosting. The models were evaluated using accuracy, precision, recall, F1-score and ROC-AUC metrics. The results indicate that ensemble models significantly outperform single learners, with Random Forest achieving the highest predictive performance. Feature importance and correlation analyses reveal that digital literacy, mobile device ownership, and educational attainment are the strongest predictors of digital adoption, while demographic variables such as age, gender, and household size exert comparatively weaker influence. The findings suggest that low adoption rates in rural communities are driven more by structural and capacity-related barriers than by resistance to technology. In conclusion, this study demonstrates the effectiveness of machine learning in modeling complex adoption behaviors and provides a scalable, data-driven framework for identifying high-impact intervention points.