Which Machine learning algorithms suits best in the material science for the problems that aims to determine the properties and functions of existing materials. Eg. typical problem of determination of band gap of solar cell materials using ML.
In material science, the approach for selecting ML involves first identifying the type of data available, determining the desired outcome or prediction, and then selecting the most appropriate algorithm based on these factors. Some common machine learning algorithms used in material science include DT, Rf, ANN, and SVM. It is also important to consider the complexity and interpretability of the algorithm, as well as the amount and quality of data available. It may be necessary to try multiple algorithms and fine-tune them for the specific application in order to achieve the best results.
Random forest, Support vector machine and Gradient Boosting Machines