Machine Learning Analysis of Literature Data on the Water Gas Shift Reaction toward Extrapolative Prediction of Novel Catalysts

Literature data based on the water gas shift (WGS) reaction have been analyzed using statistical methods based on machine learning (ML). Our ML approach, which considers elemental features as input representations rather than the catalyst compositions, was successfully applied, and new promising catalyst candidates for future research were proposed.

Automated summary

Literature data on the water gas shift reaction were analyzed using statistical methods based on machine learning. The study successfully applied a machine learning approach that considers elemental features as input representations, and proposed new promising catalyst candidates for future research.