This paper is done to suggest a much more reliable method to forecast earnings using of data mining. Most studies in this area of research are based on the statistical and econometric models, which might encounter difficulties when dealing with the nonlinearity of financial data. Although data mining has already proved to be successful in many business applications, little research has been done on integrating financial statements' analysis with its techniques. This study utilizes Multi-layer Perception (MLP) and Radial Basis Function (RBF) techniques. In this way, it examines eight features of income statement of companies listed in Tehran Stock Exchange from 2020 to 2025. The results shows that MLP and RBF approaches respectively demonstrate about 96% and 98% efficiency of classification rate .

Login Required

Please login to access the full PDF of this article.

🔑 Author Login