Title: Abnormal recognition of corporate financial data based on deep belief network

Authors: Xi Lun; Xiangyang Zhang; Yining Wang; Tian Wang

Addresses: Qinhuangdao Vocational and Technical College, Qinhuangdao 066000, China ' Qinhuangdao Vocational and Technical College, Qinhuangdao 066000, China ' Qinhuangdao Vocational and Technical College, Qinhuangdao 066000, China ' Qinhuangdao Vocational and Technical College, Qinhuangdao 066000, China

Abstract: In view of the traditional enterprise financial data exception recognition methods, such as low recognition precision and long recognition time, a deep belief network is put forward. Based on the depth of the enterprise's financial data anomaly identification method, the distributed data collection method, selection of enterprise financial data mining, and correlation analysis are adopted, according to the financial data sample information entropy, to divide the financial data flow. According to the extraction results, use the deep belief network to build a financial data anomaly recognition model. The financial data of enterprises are input into the abnormal identification model of financial data to identify the status of financial data. Experimental results show that this method has higher recognition accuracy and shorter recognition time.

Keywords: deep belief network; corporate financial data; information entropy; data stream fragment.

DOI: 10.1504/IJISE.2023.134371

International Journal of Industrial and Systems Engineering, 2023 Vol.45 No.2, pp.135 - 147

Received: 29 Jun 2021
Accepted: 18 Dec 2021

Published online: 19 Oct 2023 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article