Multi-label classification method, system and device and storage medium

Abstract

The invention discloses a multi-label classification method. First, training data is determined, then a dictionary to be learned is utilized to learn the training data to obtain the corresponding target dictionary and the sparse codes, the sparse codes are projected again to obtain the analysis dictionary, then the optimal solution of the analysis dictionary, the target dictionary, the sparse codes and the preset classifier is determined, and the corresponding optimal solution is utilized to carry out multi-label classification. Compared with the prior art, the scheme has the advantages that dictionary learning is performed on the sparse codes on the basis of one-time dictionary learning, that is, the sparse codes are projected to obtain corresponding analysis dictionaries, and then multi-label classification is performed by combining the two dictionaries. Features in the training data can be extracted more completely through two times of dictionary learning, so that the classificationresult is more accurate. The invention further provides a multi-label classification system and device and a computer readable storage medium which can also achieve the above technical effects.

Description

technical field [0001] The present invention relates to the field of label classification, and more specifically, relates to a multi-label classification method, system, device and computer-readable storage medium. Background technique [0002] In recent years, the vigorous development of Internet technology has brought earth-shaking changes to the world. Data resources have shown an exponential growth rate, with more and more data volumes and ever-changing forms. Informatization has penetrated into every corner of our lives. We are in a digital age with extremely rich information, which brings us great convenience in searching for information; however, in the face of such rich data resources, how to quickly and effectively Finding valuable information for us has become a hot spot for researchers, so various technologies for properly classifying resources have emerged as the times require. Among them, when there are multiple labels in the data set, the corresponding label c...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide a multi-label classification method, system, device and computer-readable storage medium to solve the problem of how to improve the accuracy of multi-label classification

Images