<p>In this work, the application of machine learning algorithms to the identification of constitutive model parameters is explored. Multiple machine learning algorithms are tested, considering two different datasets, one including uniaxial tensile test results, and the other including biaxial tensile test results. Of the algorithms tested, Gaussian Processes achieved the best predictive performances. Afterwards, the influence of dataset size is explored, considering both the number of materials included in the dataset, and the number of input parameters. For the latter, a feature importance methodology, namely SHAP (Shapley Additive Explanations) analysis, is applied to identify the most relevant input parameters in the datasets. The performance of the models trained with the inputs identified by the SHAP analysis is competitive for one of the datasets, which shows the potential of this type of analysis to create more efficient parameter identification methodologies.</p>