Machine Learning‐Aided Prediction and Construction of a Descriptor for Polymer Properties: A Case Study on the Lower Critical Solution Temperature of Copolymerized N‐Isopropylacrylamides

Abstract Artificial Intelligence (AI) technology has been increasingly applied in the field of chemistry in recent years. For proof of concept for the application of AI technology to polymer research, a state‐of‐the‐art machine learning (ML) technique is employed to systematically elucidate the lower critical solution temperature (LCST) of N ‐isopropylacrylamide (NIPAAm) copolymers using data collected from the literature. The information on 110 NIPAAm random copolymers is collected to extract their LCSTs and the chemical and physical parameters of the copolymers and comonomers. The ML analysis reveals that the copolymerized ratio, the elemental composition of carbon and oxygen in the comonomers, and the water solubility of the comonomers are crucial parameters affecting the LCST, and the constructed ML model successfully predicts the LCSTs of various NIPAAm copolymers. Furthermore, the genetic algorithm using symbolic regression identifies a simple and comprehensive descriptor for the LCST. This study demonstrates the usefulness of data‐driven ML techniques for polymer research with the rapid and accurate prediction of polymer properties.