<![CDATA[The rapid expansion of digital financial services in Indonesia has driven widespread adoption of e-wallet platforms, while also enabling the growth of illegal online gambling activities. This study presents a predictive framework for detecting gambling-related behavior on Indonesian e-wallet platforms using the Facebook Prophet time series forecasting model, with ARIMA employed as a comparative benchmark. By analyzing historical transaction data flagged as suspicious or potentially linked to gambling, both models are used to capture seasonal trends and detect anomalies that signal future high-risk activity. The results show that Prophet effectively anticipates spikes in gambling-related transactions, often outperforming ARIMA in capturing complex seasonal patterns. These findings highlight the potential of time series forecasting to enhance fraud detection systems by enabling proactive interventions during predicted high-risk periods. This research contributes to the field of financial fraud prevention by demonstrating the value of integrating predictive analytics particularly Prophet into efforts to combat illicit behavior within Indonesia’s evolving digital finance ecosystem.]]>
Copyrights © 2026
