A subject classification method, device, equipment and medium of data assets are disclosed. The method comprises: obtaining reference data of a data asset to be classified; obtaining a first subject
classification result of the data asset to be classified based on a first classification model according to the reference data; obtaining a second subject
classification result of the data asset to be classified based on a second classification model according to the reference data; and determining a final subject category of the data asset to be classified according to the first subject
classification result and the second subject classification result. The application inputs the reference data of the data asset to be classified into the first classification model trained according to the relationship between the
metadata of the data asset and the subject category and the second classification model trained according to the relationship between the upper data asset of the data asset and the subject category to obtain an accurate subject classification result, analyzes the subject classification result to obtain the final subject category, and realizes automatic and intelligent classification of the subject of the data asset quickly and accurately.