Abstract Chinese culture is always the eternal root of the Chinese nation and is an important language tool used in international exchanges between China and the rest of the world. As China’s status in the international arena has been enhanced in recent years, people from all over the world have become increasingly interested in the Chinese language, and more and more people are going to China to study it. Because of this, in recent years, the academic field has been researching the exchange and cooperation model of Chinese language international education, but most focus on quality rather than reality. This paper takes Chinese international education in colleges and universities as the research object and composes and summarizes the results of international scholars’ exploration of the exchange and cooperation model of Chinese international education. Based on the relevant theoretical foundation of the random forest algorithm; then, based on the previous research experience, four hypotheses are put forward on the influencing factors of the Chinese international education exchange and cooperation model, and combined with statistical theory, the correlation coefficient and visualization are used to demonstrate the size and direction of the correlation between each influencing factor and the Chinese international education exchange and cooperation model; then, the random forest algorithm is carried out from two perspectives of feature selection and weighted random forest Then, the random forest algorithm is optimised from two perspectives: feature selection and weighted random forest, and the model of Chinese international education exchange and cooperation is constructed, with the decidability coefficient (R 2) as the evaluation index of model accuracy, and the mean absolute error (MAE) and root mean square error (RMSE) as the evaluation index of model error. The four hypothetical factors (AQI) indices were analyzed for 3, 6, and 9 years respectively. The results of this study show that the index analysis of the random forest algorithm-based model of Chinese language international education exchange and cooperation is highly evaluated. In the analysis of the AQI indices for 3, 6, and 9 years, the coefficient of decidability (R 2) of the four models reached 0.84, 0.74, and 0.63, respectively, with an increase of 14.23% in the evaluation of the indicator coefficients, an average reduction of 5.77% in the mean absolute error (MAE) and an average reduction of 5.42% in the root mean square error (MAE). According to the above results, the random forest algorithm-based Chinese language international education exchange and cooperation model constructed in this paper possesses good correlation analysis, which promotes Chinese language international education exchange and cooperation and facilitates humanistic exchanges between China and other countries in the world to become more in-depth and the common development of the global economy.