Digital technologies in the accounting information system supporting decision-making processes

Purpose: The aim of the article is to characterize the possibilities of improving the accounting information system supporting decision-making processes in the enterprise with the use of selected digital technologies with particular emphasis on artificial intelligence. Design/methodology/approach: Basic research methods include critical analysis of literature. Simulation models of the stock market game using deep learning were also used. In addition, intensive computational experiments were carried out to analyze the quality of the solutions, which were determined by the proposed deep learning methods using artificial neural networks based on short-term memory (LSTM). The research presented in this article was verified by simulating the possibility of using deep learning. Findings: The results exceeded the estimates described in the literature. The average error is estimated to be less than 3% when using the LSTM network. It should therefore be assumed that other deep learning paradigms will also be an effective tool in financial systems. The results of theoretical research and numerical experiments confirmed that the impact of selected digital technologies on the improvement of the accounting information system supporting decision-making processes is significant. Practical implications: The results are the basis for formulating recommendations regarding the possibility of using the analyzed digital technologies in the accounting information system, supporting decision-making processes in the enterprise. They can also serve as an example of the digital transformation of the enterprise accounting information system. Social implications: The obtained results indicate significant opportunities to improve the accounting information system supporting decision-making processes. This situation suggests the need to implement the latest achievements of digital technologies in the accounting information system for the effective collection and processing of a growing amount of data. A clear presentation, ongoing monitoring and precise prediction of future results are the basis for making effective decisions based on precise data analysis, and not based on intuition or experience of the decision maker. Originality/value: The authenticity of the study results stems from the clear ideas for the effective use of digital technologies, in particular, deep learning with the use of artificial neural networks in the cloud to improve the accounting information system, especially in the field of estimating forecasted values.