<p>Latent heat removal during the phase changes is a quick and efficient way of heat transfer process and can be used in wide-ranging applications such as cooling systems, power generation, and cooling of heat sinks and electronic equipment. Pool boiling is a complex phenomenon and the water with different level of purity and salt contents such as distilled water, reverse osmosis (RO) water, normal water and the industrial fluids such as lubricating oil, and ethanol exhibit different thermophysical properties and different heat transfer characteristics while being boiled over a heated surface. It is essential to understand and study the effect of these properties on the pool boiling phenomena the boiling characteristics such as critical heat flux and pool boiling heat transfer coefficient values. The heated surface submerged in the respective fluid with facilities for the recording and visualization of boiling phenomenon forms the experimental arrangement for our study on the boiling phenomena. The pool boiling heat transfer coefficient and critical heat flux are measured and compared for each fluid under similar operating conditions. The difference in the values of boiling characteristics thus obtained exhibits the difference in their heat transfer behavior under pool boiling conditions. The distilled water owing to its high purity level, high heat conductivity with low boiling point yields in a higher value of critical heat flux and pool boiling heat transfer coefficient while water being similar in composition to distilled water but with dissolved impurities demonstrates rather a poor performance under pool boiling conditions and the thicker and viscous lubricating oil yields in the lowest value of critical heat flux and pool boiling heat transfer coefficient values due to its higher viscosity and lower thermal conductivity values as compared to water. Ethanol exhibits unique heat transfer behavior characterized by enhanced nucleate boiling and reduced film boiling due to its lower surface tension. In this analysis, we compared the experimental critical heat flux and pool boiling heat transfer coefficient values with artificial neural network and developed empirical correlation. In addition, Levenberg–Marquardt Machine Learning model can accurately predict the critical heat flux and pool boiling heat transfer coefficient with the mean square error (MSE) of 0.0605, 15.19 and coefficient of correlation (R²) of 0.9571 and 0.9996, respectively. The experimental results compared with artificial neural network and developed empirical correlation reveal that there is good agreement between the experimental and empirical correlation values within less than 1% and 5% deviation from the experimental data for critical heat flux and pool boiling heat transfer coefficient based on performance metrics co-efficient of correlation (R²) and mean square error (MSE) values.</p>