Research on Fraud Detection Method of Financial Data of Listed Companies Based on HMCRAN

Fubing Li and Huaqiong Duan
Additional contact information
Fubing Li: The College of Accounting and Finance, Chengdu Jincheng College, Chengdu, China
Huaqiong Duan: College of Computer and Software, Chengdu Jincheng College, Chengdu, China

International Journal of Data Warehousing and Mining (IJDWM), 2024, vol. 20, issue 1, 1-29

Abstract: This study investigates the application of deep learning technology to improve accuracy and efficiency in detecting financial fraud in listed companies. Using data from the China Economic and Financial Research Database on Main Board A-share listed companies with disclosed fraudulent behaviors, a hybrid multi-channel residual attention network (HMCRAN) was constructed, combining multi-channel residual networks and transformers. The study involved feature extraction, data preprocessing, model training, and performance evaluation, validating the superiority of the HMCRAN model for financial fraud detection. Experimental results demonstrate that the model excels in handling high-dimensional financial data and capturing long-term dependencies, significantly improving detection accuracy and robustness. The HMCRAN model outperforms traditional machine learning methods across various evaluation metrics, demonstrating its high practical application value.

Date: 2024
References: Add references at CitEc
Citations:

Downloads: (external link)
http://services.igi-global.com/resolvedoi/resolve.aspx?doi=10.4018/IJDWM.356510 (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:igg:jdwm00:v:20:y:2024:i:1:p:1-29

Access Statistics for this article

International Journal of Data Warehousing and Mining (IJDWM) is currently edited by Eric Pardede

More articles in International Journal of Data Warehousing and Mining (IJDWM) from IGI Global
Bibliographic data for series maintained by Journal Editor ().