Interpretable Machine learning based coordination motif identification scheme from X-ray absorption near-edge structure spectroscopy XANES

XANES is an important experimental method to probe the local three dimensional geometry and electronic structure of the system. The quantitative analysis of XANES data is very important to obtain the above mentioned structure. Because XANES contains a lot of information and complexity, the quantitative analysis of XANES is also a challenging task.Interpretable Machine learning(IML) approach: Shapley additive explanations(SHAP) based interpretable machine learning(ML) approach is adopted in two types of ML XANES data analysis methods respectively. These two methods are "spectrum to parameters" and "parameters to spectrum" methods.We used the TreeSHAP method to explain the mechanism of our models and of a specific prediction example. The model mechanism is explained from the physical perspective as much as possible, which expands the methodological perspective of machine learning application in XAS data analysis. The "Parameters to spectrum" model in which structural parameters are input and theoretical XANES reconstructed by machine learning algorithm are output. Taking Fe complex system as an example. The analysis results quantitatively and systematically demonstrate the model mechanism, that is, how parameter changes affect the theoretical XANES reconstructed by machine learning. This is of practical value for determining the parameter variation trend using in the next XANES fitting and improving the model. In summary, we show the application of SHAP IML in two kinds of ML models in XANES analysis field, and expand the methodological perspective of XANES quantitative analysis