Education, Science, Technology, Innovation and Life
Open Access
Sign In

Predicting Financial Fraud in Listed Companies Based on Machine Learning Models

Download as PDF

DOI: 10.23977/autml.2022.030101 | Downloads: 24 | Views: 1160

Author(s)

Jiacheng Lan 1, Peisen Li 1, Chengzeya Chen 1, Jianwen Deng 1, Sixi Gong 2

Affiliation(s)

1 Department of Math, Guangdong University of Education, Guangdong 510303, China
2 School of Electronic Information and Electrical Engineering, Huizhou University, Guangdong 516007, China

Corresponding Author

Jiacheng Lan

ABSTRACT

"Financial fraud" refers to listed companies falsifying financial statements, falsely reporting or concealing part of the financial data of the company. Based on the training data processing of missing value, exception value, standardized processing or other data processing on the training data set, among the 19 major industries in society, processing unbalanced data for 8 industries. Using SVM to select the optimal equilibrium method for each industry, and the remaining 11 industry data do not perform any equalization operations. Subsequently, The Weight, Filter, Wrapper, Embedded and other methods are used to select data characteristics, and combines the actual economic significance to obtain the final characteristics. Finally, through seven models of prediction, using AUC score as evaluation index to predict which listed companies may have fraud. On the premise of not deviating from the actual laws and practical significance, through the self-constructing function and using it to make a heat map. In this way, the similarities and differences of data indicators related to financial fraud of listed companies in different industries are obtained.

KEYWORDS

Financial fraud, Industry classification, Machine learning, Characteristic selection, Feature similarity

CITE THIS PAPER

Jiacheng Lan, Peisen Li, Chengzeya Chen, Jianwen Deng, Sixi Gong, Predicting Financial Fraud in Listed Companies Based on Machine Learning Models. Automation and Machine Learning (2022) Vol. 3: 1-7. DOI: http://dx.doi.org/10.23977/autml.2022.030101.

REFERENCES

[1] Li Hang. Statistical learning methods [M]. Beijing: Tsinghua University Press, 2012.
[2] Yuan Xianzhi, Zhou Yunpeng, Yan Chengxing, Liu Haiyang, Zeng Tu. A new approach to corporate financial fraud early warning and risk feature screening: based on artificial intelligence algorithm[C]. Proceedings of the 15th Annual China Management Conference, 2020: 709-724
[3] He Qing, Li Ning, Luo Wenjuan, Shi Zhongzhi. A review of machine learning algorithms under big data[J]. Pattern Recognition and Artificial Intelligence, 2014: 10-12
[4] Liu X.Y., Nong G.C. A comparison of several different missing value filling methods[J]. Journal of Nanning Normal College of Higher Education, 2007: 148-150
[5] Yesha. A review of missing data and its processing methods[J]. Network and Information Engineering,2017:48-50

Downloads: 1539
Visits: 65048

Sponsors, Associates, and Links


All published work is licensed under a Creative Commons Attribution 4.0 International License.

Copyright © 2016 - 2031 Clausius Scientific Press Inc. All Rights Reserved.