Due to the complexity of boiling heat transfer phenomena, the development of more accurate and widely applicable predictive models has always been an important topic in thermohydraulic research. In recent years, the application of machine learning methods to study boiling heat transfer behavior has become an emerging research direction in the field of reactor thermal hydraulics. A random forest model was developed to predict the boiling heat transfer coefficient in flow boiling inside pipes. The predictive capability of this machine learning model was analyzed and compared with the traditional Chen correlation model. The results show that within the scope of this study, the random forest model demonstrates better prediction accuracy and stability compared to the Chen correlation. The trained and optimized random forest model is able to better fit experimental data and make reliable predictions.