This dataset is compiled from a 2015 paper by X. Ma et al., in which they calculated a set of ab initio adsorption energies and electronic fingerprints of idealized bimetallic surfaces to use as the inputs for a machine-learning-augmented chemisorption model. Contains d-band and electronic features, as well as the binding energy for CO and the adsorbate-metal coupling matrix element, for 295 systems of metal alloys with one or two surface monolayers.Reference: X. Ma, Z. Li, L.E.K. Achenie, and H. Xin, (2015) Machine-Learning-Augmented Chemisorption Model for CO2 Electroreduction Catalyst Screening. J. Phys. Chem. Lett. 6. http://dx.doi.org/10.1021/acs.jpclett.5b01660