A Novel Interpretable Machine Learning Model Approach For The Prediction Of Tio2 Photocatalytic Degradation Of Air Contaminants

This article discusses the use of machine learning models to predict the photocatalytic-degradation performance of TiO in removing air-born contaminants. Four different types of models were developed and compared, using both experimental and synthetic data. Factors influencing the performance were evaluated using the SHAP method. The database used for model development consisted of 200 sets of experimental data, with eight design features and the photo-degradation rate as the response variable.