The work reported in this paper explored the use of machine learning tools to analyze quenching pool boiling data in the nucleate boiling range, near maximum heat flux range, and through the transition boiling range towards the Leidenfrost (minimum heat flux) point. It specifically explores the hypothesis that this sequence is a consequence of progressive dryout of the surface as the wall superheat increases. Machine learning tools are used with a heuristic model of the dryout parametric dependence to extract information about the magnitude of surface dryout as the superheat increases. From experimental data, the machine learning analysis provides an indication of how the dryout transition differs for different surface wetting characteristics and substrate materials. The wetting variations considered ranged from moderately wetted plain aluminum and copper surfaces to highly wetted nanostructured superhydrophilic surfaces. The data examined included aluminum and copper substrates. The results of the machine learning analysis indicate that the properties of the surface substrate can have a significant effect on the progressive surface dryout. In contrast, the surface wetting characteristics had a more limited effect for the surfaces tested. The paper concludes with an assessment of the implications of the findings for developing enhanced surfaces for boiling heat transfer performance.