Fecha de publicación: 11/04/2023 Fuente: Wipo "BigData" The invention provides a statistical big data logic model construction method and system based on data driving. The method comprises the following steps: firstly, obtaining source end data, and carrying out identification abstract processing on the source end data to obtain first data; and then performing coverage comparison on the first data and the SG-CIM logic model, and establishing a statistical big data logic model according to a comparison result. Wherein in the recognition and abstraction processing, firstly, input source end data is subjected to demand analysis, and first source data is obtained through abstraction recognition; and performing abstract extraction on the data source object, and outputting to obtain first data. Wherein abstract extraction comprises one or more of specification, semantic unification, deduplication, merging, integration and splitting. According to the method, the data and the logic relation are extracted and the logic model is constructed directly from the current situation of the data, so that the speed is high, the efficiency is high and the quality is high during implementation, meanwhile, the unified logic model of the full-energy industry is established, the uniqueness of an entity is ensured, and the real unified standard of the data model is realized.