The invention relates to the field of
computational chemistry and nano composite catalytic materials, and discloses a method for designing a
metal organic framework based on molecular fingerprints, which comprises the following steps of: 1, acquiring descriptors and adsorption performance parameters of the
metal organic framework; a second step of performing regression prediction on different molecular fingerprints by utilizing a
machine learning
algorithm based on the descriptors and the adsorption performance parameters, and taking
molecular fingerprint types as input, the
molecular fingerprint types including APFP, Estate, MACCS and PubChem; 3, according to a regression prediction effect, evaluating prediction effects of different types of molecular fingerprints, finding out a
fingerprint type with high prediction precision, and obtaining a
molecular fingerprint with high prediction precision; 4, counting excellent
fingerprint bit numbers by using the molecular fingerprints with high prediction precision; and 5, designing the
metal organic framework according to the excellent
fingerprint bit number. According to the method, the important fingerprint bit number can be accurately and quickly found, and the design of the
metal organic framework is guided.