Data classification and identification method, apparatus, computer apparatus, and readable storage medium

Abstract

The invention relates to the computer field and provides a data set classification identification method. The method comprises the following steps: acquiring a data set to be classified; Dividing thedata set to be classified into a plurality of data clusters to be classified; Judging whether the data cluster to be classified satisfies the standard defined by the sample data set; Discarding invalid data clusters to be categorized that do not meet the criteria defined by the sample data set; Determining a classification mark which is most relevant to the data cluster to be classified accordingto the characteristic information of the data cluster to be classified satisfying the standard; classifying and identifying The data clusters to be classified. A method for classifying and identifyinga data set provide by an embodiment of that present invention By dividing the data set to be classified into a plurality of data clusters to be classified, the processing efficiency is improved, andthe characteristic information of the data cluster to be classified and the characteristic information of the sample data set are utilized, so that the classification identification method of the dataset still has high robustness and accuracy depending on a small amount of sample data information.

Description

technical field [0001] The present invention relates to the field of computer technology, in particular to a data set classification and identification method, device, computer equipment and readable storage medium. Background technique [0002] The current machine learning technology is in the ascendant, especially the deep learning technology is expanding in the field of industrial application, and the ability to independently classify and label data reasonably through learning is one of the foundations of machine deep learning technology. [0003] In the existing field of text classification, the open source corpus content published by scientific research websites is mainly used for scientific research purposes, but cannot train the corresponding classification models required by the industrial environment, but can be used to train the corresponding classification models required by the industrial environment for commercial use. Most machine learning algorithms rely on ma...

AI Technical Summary

Problems solved by technology

[0003] In the existing field of text classification, the open source corpus content published by scientific research websites is mainly used for scientific research purposes, but cannot train the corresponding classification models required by the industrial environment, but can be used to train the corresponding classification models required by the industrial environment for commercial use. Most machine learning algorithms rely on marked databases, but marked databases are built on the basis of a large amount of manpower collected and marked manually. The larger the marked database, the higher the labor cost. Smaller labeled databases cannot guarantee sufficient accuracy and robustness of the trained classification model

[0004] It can be seen that in the existing technology, the classification model corresponding to the industrial environment has technical defects such as high cost, low accuracy of classification and identification, and low robustness, which cannot meet the needs of current machine classification and identification.

Images