The objective of this study is to estimate the energy dissipation capacity of corroded reinforced concrete columns using the machine learning-based M5P decision tree method. The dataset was obtained from full-scale experiments on corroded reinforced concrete columns conducted by Yalciner and Kumbasaroglu (2020) [1]. In this experimental program, the cyclic behavior of reinforced concrete columns was investigated by considering different concrete compressive strengths, longitudinal reinforcement corrosion levels, stirrup corrosion levels, and axial load ratios. In this paper, concrete compressive strength, longitudinal reinforcement corrosion level, and stirrup corrosion level were used as model inputs; energy dissipation capacity was selected as the target variable. The modeling process was performed using the M5P algorithm in Weka software, and model performance was evaluated using a 10-fold cross-validation method. According to the results, the M5P model divided the dataset into two regions based on a threshold value of 32.63 MPa for concrete compressive strength. For columns with concrete compressive strength of 32.63 MPa or lower, the energy dissipation capacity was predicted as a linear function of concrete compressive strength; in the higher strength region, the model produced a constant predicted value. As a result of 10-fold cross-validation, the correlation coefficient was found to be 0.5045, the mean absolute error 8.4973, and the root mean square error 10.1181. The results indicate that, although the M5P model produces interpretable equations, its predictive performance remains at a moderate-to-low level given the current variables and limited dataset.