This paper empirically examines the relationship between economic freedom and income inequality by using a machine-learning constructed economic freedom indicator that addresses three major methodological problems in the currently popular economic freedom indices. This paper selects the optimal Support Vector Regression (SVR) model with Radial Basis Function (RBF) kernel to construct the new economic freedom indicator, after comparing four different kernel types. The empirical results show that there exists a robust invert-U shaped relationship between economic freedom and income inequality.