The invention discloses a regional industrial structure intelligent
analysis method based on enterprise
big data mining, and relates to the technical field of enterprise
big data mining and governance, and the method comprises the steps: obtaining an enterprise registration address and activity occurrence place
record, and constructing a weighted registration
point set and a weighted activity
point set; respectively calculating statistical parameters to construct a registration standard deviation
ellipse and an activity standard deviation
ellipse; based on the area difference of the two types of ellipses, the distribution difference of the active points inside and outside the ellipses and the center
offset distance, synthesizing and calculating the double-
ellipse hollowing
fingerprint intensity; a hollowing significant industry set is determined through statistical baseline screening; and finally, adaptively selecting an activity standard deviation ellipse or a registration standard deviation ellipse as an industrial space range according to a
screening result, and generating a regional industrial portrait. According to the method, the spatial
dislocation of registration and operation is effectively quantified, and the accuracy of industrial feature analysis is improved.