Answers are grounded in this paper and cite their sources.
In this study, through the use of machine learning models and BP neural network, a large amount of financial data is deeply mined and analysed. The research results show that the constructed model improves the financial fraud recognition accuracy to 0.94, which can achieve the discrimination of financial fraudulent account behaviours at a high accuracy rate, provide strong support for financial institutions, and effectively prevent financial fraud risks. This study aims to solve the problem of high variability and concealment of financial fraud, different from the direction of departure of the more significant results of fraud identification for financial transactions, but focuses on the portrayal and analysis of the characteristics of the financial fraud account, so as to achieve the efficient identification of the source of financial fraud decision-making.
No takes yet. Share an insight, caveat, or question.
Geng et al. (2024) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: