The dynamic landscape of Financial Technology (Fintech) enterprises has catalyzed transformative innovations, yet concurrently spurred challenges in fraud detection and mitigation. This study investigates the integration of Business Intelligence (BI) strategies in addressing sustainable fraud mitigation within the Fintech domain. Grounded in the problem statement of class imbalance, our methodology leverages Adaptive Boosting (AdaBoost) and Synthetic Minority Over-sampling Technique (SMOTE) to augment fraud detection models. Through empirical analyses, including feature importance assessments and ROC curve evaluations, this research reveals the efficacy of BI-driven approaches in enhancing predictive accuracy, discerning critical fraud indicators, and mitigating the challenges posed by class imbalance. Visual representations, such as correlation maps, elucidate intricate relationships among variables, offering crucial insights for model refinement. The study concludes by emphasizing the pivotal role of BI-integrated strategies in fostering resilient fraud mitigation frameworks within Fintech enterprises, paving the way for continued innovation and trust-building in the evolving landscape of financial technologies.