 This paper presents a comprehensive data-driven modeling experiment. It proposes six DDM techniques, multiple linear regression, and naive models for comparison. The paper then creates 12 different realizations from each dataset, containing three subsets, training, cross-validation, and testing. These subsets are used to evaluate the performance of the DDM techniques and compare them against the baseline. Finally, the paper summarizes its findings and concludes that the best performing DDM technique was support vector machines, followed by genetic programming and evolutionary polynomial regression. This article was authored by A.L. Shorbighi, G. Corzo, S. Srinivasulu, and others.