Machine Learning for the Design and Optimization of Gas-Sensing Materials: A Review

Gas-sensitive materials are widely used in fields such as food testing, environmental monitoring, and medical diagnosis. However, their research and development have long been constrained by the high cost and low efficiency of traditional trial-and-error methods. Machine learning (ML) has significantly accelerated the research and development process of gas-sensitive materials through data-driven material design, process optimization and performance prediction. This paper reviews the key technologies and application progress of ML in the preparation of gas-sensitive materials: Firstly, it compares and analyzes the characteristics and applicable scenarios of traditional and deep ML algorithms; Secondly, a detailed discussion was conducted on the general process of ML in the preparation of gas-sensitive materials, as well as specific application cases of ML in material design and screening, optimization of the preparation process, characterization analysis and performance prediction. Finally, the current research progress is summarized and future directions such as automated closed-loop optimization and edge intelligent sensing are prospected.