Automatic correlation method facing multivariate data

Abstract

The invention discloses an automatic correlation method facing multivariate data. The method includes selecting multivariate data, extracting time attribute of each data, conducting automatic supplementation with reference to the text for the data with insufficient time attribute, extracting the geometrical center position and the covering range of each data and converting the geometrical center position and the covering range to a unified oval coordinate system, adding attribute labels for the data, automatically adding the attribute labels for the data according to the source, the category, the class and the resolution of the data, wherein the data with identical labels have the implicit correlation; utilizing the multi-attribute information to excavate the correlation between the data, meanwhile considering the time attribute, the position attribute and the label information of the data, utilizing the improved distance to excavate the correlation between the data through the Chinese restaurant process, storing the correlation between the data as acknowledged information for follow-up automatic correlation. The method has the advantages of fully considering various attribute information to conduct data correlation to effectively utilize a large amount of information.

Description

technical field [0001] The invention belongs to the technical field of multivariate data automatic association, and relates to an automatic association method for multivariate data. Background technique [0002] With the development of aerospace technology and sensor technology, more and more remote sensing data can be obtained. How to automatically organize and manage these massive remote sensing data is the prerequisite for effective use. At the same time, these massive data come from different sensors, and the attribute elements of each sensor are also different, so the obtained data is also diverse. Multivariate data has different structural information, how to automatically organize and associate these data is the prerequisite for the effective use of information. [0003] At present, the association of multivariate data is mostly automatically associated based on specific attribute information. This association does not jointly consider multiple attribute information,...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide an automatic association method for multivariate data, which solves the problem that the existing association method does not jointly consider multiple attribute information, and it is easy to cause too many association relationships, reduce the effectiveness of association, and solve the automatic association of massive remote sensing data question

Images