The present study aims to optimize the bank financing process of small and medium-sized enterprises (SMEs) using data mining algorithms and artificial intelligence. In the present era, SMEs play a vital role in the dynamics of the economy, but access to efficient financial resources is their main challenge. Traditional bank financing methods do not meet the dynamic needs of SMEs due to time-consuming processes and inadequate risk assessments. This research is of an applied type with a mixed approach (qualitative-quantitative) and an integrated strategy and aims to provide a smart and efficient model to facilitate SMEs' access to financial resources and reduce banks' credit risk. In the qualitative part, the meta-synthesis method was used to extract key features affecting SME financing, and in the quantitative part, by analyzing real data of Tejarat Bank's credit granting to 1073 SMEs over a five-year period, modeling was done with machine learning algorithms (linear regression, decision tree, k-nearest neighbor, support vector machine, and artificial neural networks). The results of the qualitative part led to the identification of four main dimensions affecting SME financing, including: business characteristics, business strategy, financial and credit status, and external factors. In the quantitative part, different algorithms were evaluated using these features and their combined categories. The findings showed that the artificial neural network (ANN) algorithm, using all extracted features, has the highest accuracy (95.75%) and stability in predicting the success of SME financing. Also, the characteristics of the financial and credit status of the business were identified as the most important categories of characteristics in the financing decision-making process. Based on these findings, the final conceptual model of the research explains the relative importance of different categories of characteristics in the decision-making of Tejarat Bank for financing SMEs. This research conclusively proves the efficiency and effectiveness of machine learning algorithms, especially artificial neural networks, in optimizing the process of bank financing of SMEs and shows their potential in improving the credit assessment processes, reducing credit risks, facilitating SMEs' access to financial resources and increasing the efficiency and reducing the costs of banks.