Title: An enterprise financial data leakage risk prediction based on ARIMA-SVM combination model
Authors: Qian Cao
Addresses: Department of Economics and Management, Jiangsu Vocational Institute of Architectural Technology, Xuzhou 221116, China
Abstract: Aiming at the problems of low prediction accuracy and long prediction time in traditional methods, an enterprise financial data leakage risk prediction method based on ARIMA-SVM combination model is proposed. According to the financial data security risks existing at all levels of the enterprise, the enterprise financial data leakage risk prediction index systemis built and the prediction indexes, including the two first-class indexes of application security and system security, the second-class indexes such as foreign cooperation security, and the third-class indexes such as the identification of sensitive data are obtained. Empirical mode decomposition is used to remove the noise data of prediction index, and the data after noise removal is input into ARIMA-SVM combination model, and the output of the model is the prediction result of data leakage risk. The simulation results show that the prediction accuracy of the proposed method is between 95%~100%, and the prediction time is within 16 s.
Keywords: ARIMA-SVM combined model; enterprise financial data; risk of leakage; empirical mode decomposition method.
DOI: 10.1504/IJASS.2023.134358
International Journal of Applied Systemic Studies, 2023 Vol.10 No.3, pp.169 - 181
Received: 07 Dec 2021
Accepted: 14 May 2022
Published online: 19 Oct 2023 *