Ran Li
Langfang Polytechnic College,Langfang, 065001,China

DOI:https://doi.org/10.5912/jcb1037


Abstract:

Aiming at the problems of high damping coefficient and long aggregation response time of traditional enterprise financial supervision data aggregation model, an enterprise financial supervision data aggregation model based on BP neural network is designed. In order to effectively gather enterprise financial supervision data, an enterprise model of financial supervision data is proposed based on concurrent workflow. Using the basic financial data, construct the tree workflow, use the financial data summary server tree, construct the multi-level data aggregation mechanism, compress the enterprise financial supervision data based on BP neural network, and realize the model through the tree workflow and multi-level data aggregation mechanism. The experimental results show that the aggregation damping coefficient and response time of the design model are significantly lower than those of the control group, which can solve the problems of high damping coefficient and long aggregation response time of the traditional enterprise financial supervision data aggregation model. The heart of enterprise financial distress theory is financial distress prediction, which is a critical link in business risk management. To begin with, the evolutionary algorithm is used to optimise the back-propagation (BP) neural network model in light of the present global economic downturn and the ongoing perfection of artificial intelligence technology (GA). Thus, the BP neural network model's weakness of being sluggish to convergence and easily caught in local optimum solutions may be remedied. The GA-BP neural network model is then trained and tested using financial distress data from Chinese listed companies. Results show that the GA-BP neural network model is far more accurate and stable when predicting financial hardship than the baseline model. A successful test model for the automated detection and early warning of company financial trouble has been developed in this work, as well as new ideas and methodologies for the use of artificial intelligence in financial accounting.