WCSE

Financial Risk Early Warning Model of Industrial Listed Enterprises Based on Deep Learning

Abstract— In this paper, the public financial indicators of 156 listed industrial enterprises in China Stock Market in T-2 year(2017) are used to predict the company financial status in T year(2019). Based on the feature extraction of Random Forest, a double LSTM financial risk early warning model is built through Keras framework. Machine learning algorithms including LR, SVM, KNN and NBC are set as baseline models. To reduce the influence of unbalanced data on the model, the G-mean is introduced as the comprehensive measure of the model. The result shows that G-mean of RF-LSTM on the test set is much higher than that of machine learning model which verifies the practicability of the RF- LSTM model.

Index Terms— LSTM, random forest, G-mean, financial risk

Xiaohui Yu, Shihong Chen
School of Information Science and Technology, Guangdong University of Foreign Studies, Guangzhou, CHINA
Shijie Chen
Graduate School of Business, National University of Malaysia, Bangi, MALAYSIA
Jiaying Tan
School of Accounting, Guangdong University of Foreign Studies, Guangzhou, CHINA

[Download]

Cite: Xiaohui Yu, Shihong Chen, Shijie Chen, Jiaying Tan, "Financial Risk Early Warning Model of Industrial Listed Enterprises Based on Deep Learning, " Proceedings of 2022 the 12th International Workshop on Computer Science and Engineering (WCSE 2022), pp. 1-6, June 24-27, 2022.