Hello! Great article. I wanted to understand a few things better, though. You say that the simplified plot highlights correlations with the target variable. In the ABS_SHAP function, though, you compute correlation values between shaps and feature values. Do you have a mathematical /theoretical explanation of this procedure?