In this study, machine learning algorithms and statistical learning methods—including Linear Regression, Simple Linear Regression, Additive Regression, $M_5$ Rules, and Gaussian Process—were employed to predict the Nusselt number in plate heat exchangers. A comprehensive database consisting of experimental data from various sources was compiled and utilized. The input parameters included the chevron angle, aspect ratio of the corrugated profile, surface enlargement factor, and Reynolds number, while the Nusselt number was considered the output variable. The results indicated high accuracy of the applied models, with the $M_5$ Rules and Additive Regression methods achieving the highest correlation coefficient on the training data. These methods were also identified as the most accurate models on the test data, exhibiting the highest correlation coefficient and the lowest error. The Mean Absolute Percentage Error and Root Mean Square Error values for this model on the test data were calculated at the lowest levels. The findings of this study demonstrate that the M5 Rules method not only offers very high predictive accuracy but also, due to its transparent structure and interpretability, serves as a reliable tool for modeling heat transfer phenomena in plate heat exchangers across a wide range of operating conditions.