A novel heuristic-based selective ensemble prediction method for digital financial fraud risk

In the era of artificial intelligence and big data, financial fraud is inevitable. This paper proposes ENKMRH, a novel selective ensemble prediction method for combatting fraud risk in digital financial systems, designed to adapt to complex, high-dimensional, and non-linear financial data. First, we innovatively apply K-means++ for the diversified selection of well-performing base learners and employ distance-based selection to preliminarily select partial learners with better overall performance. This approach conserves computing resources and improves the generalisation ability and stability of the ensemble system. Second, we introduce chaos initialisation, an improved position update, an escape energy strategy based on biological principles, and a refractive inverse learning mechanism into the standard Harris Hawks optimisation algorithm (RILHHO). RILHHO is designed to provide an efficient and precise selection strategy for the selective ensemble prediction of digital financial fraud risk. Finally, we apply the proposed model to three typical real-world problems in digital financial fraud: internet consumer credit fraud, online lending fraud, and money laundering risk prediction. The experimental results demonstrate that ENKMRH outperforms other state-of-the-art basic techniques and ensemble learning models, achieving the highest accuracy rates of 81.39%, 88.68%, and 93.80% across three financial fraud datasets. The research findings offer crucial guidance to financial practitioners, aiding in investment decisions and bolstering financial stability and security. Furthermore, they enhance institutions’ risk management capabilities, fostering sustainable growth and prosperity. This paper intertwines financial risk management with the domains of technology and engineering management, utilising advanced algorithms and data analytics techniques to tackle modern challenges in digital financial systems.
Managerial relevance statement—The managerial significance of this work lies in its profound implications for enhancing fraud risk prediction in real-world financial settings, particularly in the digital realm. By addressing the intricate challenges posed by complex, high-dimensional, and non-linear transaction data, this work offers a practical solution to fraud detection. This study not only enhances the accuracy, generalisation ability, and stability of fraud risk prediction but also provides a deeper understanding of the underlying data features crucial for fraud risk prediction. Through the utilisation of advanced data analytics techniques, financial institutions can gain insight into patterns and anomalies within transaction data, enabling more proactive and targeted fraud detection strategies, thereby achieving effective risk management. Moreover, the adoption of ENKMRH enables financial practitioners to better address a wide range of fraud scenarios, including internet consumer credit fraud, online lending fraud, and money laundering risks. By empowering institutions with a robust fraud detection tool, ENKMRH contributes directly to reducing bad debt losses and reinforcing overall stability and security in the digital age. In essence, through its innovative methodologies and insights into data-driven decision-making, this study offers tangible contributions to enhancing risk management practices and ultimately safeguarding the integrity of financial systems.