Loading [MathJax]/extensions/MathMenu.js
Supporting Financial Inclusion with Digital Financial Services: Comparative Analysis of Machine Learning Models | IEEE Conference Publication | IEEE Xplore

Supporting Financial Inclusion with Digital Financial Services: Comparative Analysis of Machine Learning Models


Abstract:

While Malaysia has made significant progress in promoting financial inclusion, limited access to banking services due to physical proximity and a limited branch network c...Show More

Abstract:

While Malaysia has made significant progress in promoting financial inclusion, limited access to banking services due to physical proximity and a limited branch network continues to hinder its full realization. Digital financial services can facilitate the promotion of financial inclusion as they offer enhanced accessibility, convenience, and affordability to individuals who face limited availability of traditional financial services. The objective of this research is to examine whether digital financial services can drive financial inclusion in Malaysia by using machine learning models. Using the Global Findex 2021 Database, this study employs supervised machine learning models to predict the roles of digital financial services in influencing financial inclusion. Based on a comparison among the methods in supervised learning (linear regression, logistic regression, decision trees, random forest, support vector machines, gradient boosting, naive Bayes, K-nearest neighbors, and neural networks), the random forest model demonstrates the best performance in using digital financial services factors to predict financial inclusion. The random forest model suggests that making or receiving digital payments, buying something online and making bill payments online using the internet are the most influential factors driving financial inclusion. The implication of this study suggests that governments, financial institutions, and service providers could prioritize these elements to effectively promote financial inclusion. The novelty of this study lies in the use of machine learning models to accurately predict the variables that explain financial inclusion, providing the capability to make precise predictions for new, previously unseen factors based on the knowledge acquired from the training data.
Date of Conference: 12-14 September 2023
Date Added to IEEE Xplore: 27 October 2023
ISBN Information:
Conference Location: Kota Kinabalu, Malaysia

Contact IEEE to Subscribe

References

References is not available for this document.