rukayah/Africa-Financial-Inclusion-Predictions

# Africa Financial Inclusion (bank account) Prediction ## Introduction Financial inclusion remains one of the main obstacles to economic and human development in Africa. For example, across Kenya, Rwanda, Tanzania, and Uganda,only 9.1 million adults (or 14% of adults) have access to or use a commercial bank account. Access to bank accounts has long been thought of as a sign of financial inclusion. Banks continue to be crucial in facilitating access to financial services, despite the widespread use of mobile money in Africa and the expansion of creative fintech solutions. ## Data Source The data set is rather big, with 23,524 records for the train data and 10,086 records for the tests. Given that many African nations are unbanked or not financially integrated, it would be preferable to forecast outcomes given certain characteristics in order to identify reliable indicators that would allow residents of the four nations we have selected to choose to be financially involved. Financial Inclusion in Afri… ## Key Performance Metrics with Machine Learning I want to predict the likelihood of the person having a bank account or not (Yes = 1, No = 0), for each unique id in the test dataset . I would train my model on 70% of the data and test model on the final 30% of the data, across four East African countries - Kenya, Rwanda, Tanzania, and Uganda. This would be done with machine learning models to ensure we predicts accurately whether a person would likely be financial included. ## Metrics Prediction the bank account is a classification machine learning problem. Therefore, the following metrics that are useful for classification problems are taken in account. * Accuracy * Connfusion matrix * f1_score I used the XGBClassifier from xgboost with an accuracy of 0.89. ## Insights from the data ## Future Scope * It would be excellent to understand how much features such as location_ty …