ARTIFICIAL INTELLIGENCE AND FINANCIAL INCLUSION IN SELECTED EAST AFRICAN COUNTRIES

Abstract

The study addresses the use of artificial intelligence in predicting financial inclusion in East Africa, particularly the lack of bank accounts which negatively impact development and livelihood. To address this issue, the study proposes the use of machine learning techniques to predict individuals who own bank accounts and those who do not. The study employed various machine learning algorithms, including Logistic Regression, Naive Bayes, Random Forest Classifier, Decision Tree, Gradient Boosting, XGBoost, SVM, and K-Nearest Neighbours, to predict financial inclusion in East Africa. A public dataset obtained from Kaggle was used to predict whether an individual has a bank account or not. The study built a machine learning model that predicts financial inclusion in East Africa using individuals' demographic and economic information. The Gradient Boosting model outperformed other models, with a mean accuracy score of 0.89. Feature importance analysis revealed that Level of Education was the most significant predictor of financial inclusion, followed by Type of Job and the relationship with head. The study highlights the most essential factors in promoting financial inclusion in East Africa and provides insights for policymakers and financial institutions to improve access to financial services. Based on the results of the study, it was recommended that policymakers and financial institutions in East Africa should focus on improving financial inclusion for individuals with a higher level of education, especially those with vocational or specialized training and secondary education. Additionally, efforts should be made to improve access to cell phone technology, particularly in urban areas.