Title: Abnormal detection method of accounting data based on information extraction technology

Authors: Le Li

Addresses: Department of Economic Management, Henan Polytechnic Institute, Nanyang, 473000, China

Abstract: In order to overcome the problems of low detection accuracy and efficiency of traditional abnormal detection methods for accounting data, this paper proposes a new abnormal detection method of accounting data based on information extraction technology. This method uses the information extraction technology based on XML to extract the accounting data, and eliminates the redundant data in the accounting data through the redundancy strategy of erasure code and full backup and the dynamic copy management optimisation model. Using the processed accounting, the principal component analysis method is used to realise the abnormal detection of accounting data. The experimental results show that the proposed method has high detection accuracy, the highest detection accuracy is 0.98, and the highest detection efficiency can reach 97%, which can meet the requirements of abnormal detection of accounting data.

Keywords: information extraction technology; redundant data elimination; accounting data; principal component analysis; data's abnormal detection.

DOI: 10.1504/IJICT.2022.124337

International Journal of Information and Communication Technology, 2022 Vol.21 No.1, pp.63 - 78

Received: 09 Sep 2020
Accepted: 26 Oct 2020
Published online: 22 Jul 2022 *

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