Can the use of machine learning techniques on financial inclusion dataset in Eswatini be a game changer?

DeepLearningIndaba 2025 Workshop AIBF Submission9 Authors

Keywords: Financial Inclusion, Machine learning, Eswatini, linear regression, support vector machine

Abstract: Lack of data-driven intelligent applications for re-engineering financial inclusion and for determining an individual’s or a country's Financial Inclusion (FI) status in real-time is a barrier. To harness the power of machine learning predictive models to build a sustainable society and inclusive economy in the Kingdom of Eswatini, it is imperative to align the Eswatini financial inclusion strategy with the financial inclusion framework based on the recommendation of Alliance for financial inclusion (AFI). It is when the alignment is properly done that we can develop a reliable and dependable financial inclusion (FI) scheme for citizens and businesses to improve access to financial services and contribute to economic growth. Most households in local communities do not have a bank account so the government needs to come up with a policy to encourage a cashless society that will also give room for credit accessibility and use of the financial institution. Machine learning algorithms are a game changer because it can tremendously assist policymakers, individuals, countries and financial service providers with effective ways of providing and managing financial services across different population groups to ensure inclusive participation, economic growth and financial inclusion. This can also help regulators make informed decisions on financial inclusion and monitor the financial inclusiveness of various households in the four regions of Eswatini.

Submission Number: 9