Article: Intelligent recognition of financial fraud based on CART decision tree Journal: International Journal of Information and Communication Technology (IJICT) 2025 Vol.26 No.11 pp.1 - 20 Abstract: This paper proposes an intelligent recognition model of financial fraud based on classification and regression tree (CART) decision tree, which aims to improve the recognition rate of financial fraud and provide a preliminary reference for other industries to use non-financial information for fraud recognition. The decision tree model adopted is tuning Iterative Dichotomiser 3 (ID3) algorithm and CART algorithm, and optimises the decision tree parameters by particle swarm to avoid the occurrence of over-fitting. It is found that the area under curve (AUC) of CART tree recognition method is significantly higher than that of random forest (RF) and neural network recognition methods, reaching 70%, which has a good recognition effect. It can be seen that parameter combination search can make the accuracy of CART decision tree model achieve the best effect, and has a positive effect on improving the intelligent recognition effect of financial fraud behaviour. Inderscience Publishers - linking academia, business and industry through research

Title: Intelligent recognition of financial fraud based on CART decision tree

Authors: Guiyun Chen

Addresses: School of Management, Changsha Medical University, Leifeng Avenue, Wangcheng District, Changsha, Hunan, 410219, China

Abstract: This paper proposes an intelligent recognition model of financial fraud based on classification and regression tree (CART) decision tree, which aims to improve the recognition rate of financial fraud and provide a preliminary reference for other industries to use non-financial information for fraud recognition. The decision tree model adopted is tuning Iterative Dichotomiser 3 (ID3) algorithm and CART algorithm, and optimises the decision tree parameters by particle swarm to avoid the occurrence of over-fitting. It is found that the area under curve (AUC) of CART tree recognition method is significantly higher than that of random forest (RF) and neural network recognition methods, reaching 70%, which has a good recognition effect. It can be seen that parameter combination search can make the accuracy of CART decision tree model achieve the best effect, and has a positive effect on improving the intelligent recognition effect of financial fraud behaviour.

Keywords: CART decision tree; finance; fraud; intelligent recognition.

DOI: 10.1504/IJICT.2025.146100

International Journal of Information and Communication Technology, 2025 Vol.26 No.11, pp.1 - 20

Received: 16 Jan 2025
Accepted: 26 Feb 2025
Published online: 06 May 2025 *

Title: Intelligent recognition of financial fraud based on CART decision tree

Keep up-to-date