Title: Design of a financial reporting management generation system based on Bi-LSTM model and MultiWord-Embedding method

Authors: Yu Wang

Addresses: Henan Institute of Economics and Trade, Zhengzhou, 450000, China

Abstract: A text classification-based financial report management generation system is proposed to address the issues of slow data collection and low data integration efficiency in existing financial report management generation systems. The overall system focuses on financial data collection and classification, with a processor, human-machine interaction module, financial data collection and classification module, financial data storage module, financial data backup module, and financial statement generation module. In the financial data collection and classification module, an improved text classification method of Bi-LSTM+MultiWord-Embedding with lexical attention mechanism is proposed to address the problem of insufficient feature extraction ability of financial text data in traditional text classification methods by combining the features of a financial text. The experimental results show that the accuracy and recall of the system in this paper reach 0.93 and 0.91 respectively, which can achieve accurate and stable financial text data classification.

Keywords: deep learning; text classification; neural networks; financial report management generation system.

DOI: 10.1504/IJDMB.2024.137747

International Journal of Data Mining and Bioinformatics, 2024 Vol.28 No.2, pp.156 - 167

Received: 20 Apr 2023
Accepted: 15 Sep 2023
Published online: 04 Apr 2024 *

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