Prediction of the first 2+ states properties for atomic nuclei using Light Gradient Boosting Machine

We explored the feasibility of using machine learning (ML) methods to calculate the energy of the first 2+ state of atomic nuclei. Based on the machine learning model of LightGBM, we have successfully predicted the excitation energy of the first 2+ state of the majority of even even nuclei from He to Rf in a reasonable manner. This dataset contains computational results and corresponding four sets of input physical features. This dataset is provided for free and can be used as a sample for ML research. Detailed information can be found in related articles.