As a fundamental part of the economy, the corporate system plays a vital role in value creation and economic growth. In volatile economic conditions, predicting and managing the financial performance of companies is particularly important. Focusing on the listed companies of Iran and Iraq, this research has investigated the effect of corporate governance, auditing and accounting quality of the public sector on financial performance using financial and economic data from 2008 to 2023. The models used include Decision Tree, Gradient Reinforcement, Support Vector Machine (SVM), Support Vector Regression (SVR), Artificial Neural Network (ANN), Multilayer Perceptron (MLP), Gaussian Neural Network (RBF), Probabilistic Neural Network (PNN). Recurrent Neural Network (RNN) and Expected Optimization Algorithm. The results of the research showed that in Iran, the decision tree model showed the most efficiency in predicting the financial performance of companies with an accuracy of 81.81%. This shows the importance of non-linear structure in the relationships between the variables of corporate governance, audit quality and public sector accounting with financial performance. In the area of neural networks, the recurrent neural network model with a root means square error of 1.701 and the multilayer perceptron model with an average absolute error percentage of 0.8115 had superior performance, indicating the importance of considering temporal dependencies in financial data. Support vector machine models and the expectation optimization algorithm with an accuracy of 74.37% also produced acceptable results. A similar pattern was observed for Iraq, but with minor differences. The adversarial learning model, with a mean square error of 0.0992, performed better in Iran, which can be attributed to the difference in corporate structure and level of financial market development between the two countries. According to the results of the research, the key variables affecting the financial performance of companies in Iran and Iraq, including the structure of the board of directors, the independence of the auditor, financial transparency, the quality of information disclosure and the implementation of public sector accounting standards were recognized as the most important factors. The results also emphasise the importance of paying attention to the complex interaction between the variables of corporate governance, audit quality and public sector accounting with financial performance, especially during periods of economic recession. Finally, this research highlights the need to develop performance assessment systems based on machine learning and neural network models, which can contribute to the timely identification of poorly performing companies and preventive intervention, ultimately leading to the strengthening of financial stability and sustainable economic growth.