Recent advances in knowledge discovery for heterogeneous catalysis using machine learning

ABSTRACT The use of machine learning (ML) in catalysis has been significantly increased in recent years due to the astonishing developments in data processing technologies and the accumulation of a large amount of data in published literature and databases. The data generated in house or extracted from external sources have been analyzed using various ML techniques to see patterns, develop models for prediction and deduce heuristic rules for the future. This communication aims to review the works involving knowledge discovery in catalysis using ML techniques; the basic principles, common tools and implementation of ML in catalysis are also summarized. Abbreviations: ANN: Artificial neural network; ASLA: Atomistic structure learning algorithm; CatApp: A web application heterogeneous catalysis; CSD: Cambridge Structural Database; co-pre: Co-precipitation; Cx: Fraction of curvature; DFT: Density functional theory; DT: Decision tree; ∆ECO: CO adsorption energy; Fx: Fraction of facets; MBTR: Many-body tensor representation; ML: Machine learning; MOF: Metal-organic framework; Nx: Number of atoms; PC: Polymerized complex; Rx: Radius; R2: Coefficient of determination; RMSE: Root mean square error; RSM: Response surface methodology; SG: Sol-gel; SISSO: Sure independence screening and sparsifying operator; SIMELS: Simplified molecular-input line-entry system; SOAP: Smooth overlap of atomic positions; SSR: Solid-state reaction; T: Temperature; t: Time; τ: Atomic deposition rate; WIPO: World Intellectual Property Organization; WOS: Web of Science; XANES: X-ray absorption near-edge structure