The electronic work function is a simple and basic parameter, which can well connect the properties of materials with their electronic behaviors. Three kinds of elastic modulus are used to predict the electronic work function of pure metals. Different machine learning methods are used to establish regression models, which are based on linear regression, decision tree, ensemble algorithm, support vector machine and neural network. The cross-validation method is used to improve the prediction accuracy and generalization ability, and the fit of the model is evaluated by several indicators. The results show that the fit of ensemble algorithm is better than other machine learning methods. Through this experiment, we can see that the model based on machine learning will not only become an accurate prediction of material properties, but also become a particularly useful tool to accelerate the design of alloy materials.
₹10000 (INR)
IEEE-2021