Improved Machine Learning Algorithms for Fraud Detection in Fintech Companies

The rapid adoption of mobile money is due to the penetration of smart devices and the reduced cost of computing power. Today, many rural dwellers use mobile smart devices (Ofcom, 2022). A financial solution that creates a transaction system through mobile smart devices is instrumental in closing the financial inclusion gap. Mobile money offers digital financial service providers access to an untapped market that is profitable. There has been a shift from traditional banking to digital banking services (mobile money). This is due to a rapid increase in mobile phone penetration and reliance on domestic remittances in many households (Mauree and Kohli, 2013). With the broad internet usage, modern technologies, and the current growth of digital financial service providers, there is a need for financial institutions to mitigate against financial crimes and fraudulent activities (Wewege et al., 2020). The sporadic expansion of digital transformation has made digital financial services providers vulnerable to cyber risks. The continuous partnerships with third parties, technology providers and fintech companies have heightened their vulnerability to cyber-attacks (Tsys, 2017). More people are prioritizing their delicate personal data against hacking, identity theft, cyberattacks and money laundering. Digital banks with lesser security infrastructure and small fintech companies are more prone to financial fraud and cyber risks. Hence, digital financial service providers and fintech companies should continuously update their security infrastructure (Wewege et al., 2020). This study focuses on how fintech companies can maximize improved machine learning algorithms for fraud detection.