Text classification method and device and computer equipment

Abstract

The invention discloses a text classification method and device and computer equipment, and the method comprises the steps: obtaining a to-be-classified text and a preset class label vector; performing content coding processing on the to-be-classified text based on a coding channel of the text classification model to obtain a content coding vector; determining a to-be-decoded vector according to the preset category label vector and the content coding vector; decoding the to-be-decoded vector based on a decoding channel of a text classification model, the target condition vector and a target category label vector output by the decoding channel at the previous moment to obtain a category label of the to-be-classified text, wherein the target condition vector is a vector determined accordingto a content coding vector and a hidden layer state vector of a decoding channel at the previous moment. According to the method, the capability of capturing the relationship between the to-be-classified text and the category labels and the relationship between the category labels is improved, so that the output category labels of the to-be-classified text are more accurate, and the text classification accuracy is improved.

Description

technical field [0001] The invention relates to the technical field of computers, in particular to a text classification method, device and computer equipment. Background technique [0002] In related technologies, texts that may belong to multiple categories at the same time are called multi-label texts. With the development of artificial intelligence technology, multi-label text classification methods based on machine learning are widely used. [0003] However, in the prior art, the multi-label task is generally divided into multiple single-label binary classification tasks, and the relationship between the text to be classified and the category label is used for classification. Although this classification method can capture the relationship between the text to be classified and the category label relationship, but ignores the relationship between class labels and class labels, resulting in poor accuracy of text classification. Contents of the invention [0004] In ord...

AI Technical Summary

Problems solved by technology

[0003] However, in the prior art, the multi-label task is generally divided into multiple single-label binary classification tasks, and the relationship between the text to be classified and the category label is used for classification. Although this classification method can capture the relationship between the text to be classified and the category label relationship, but ignores the relationship between category labels and category labels, resulting in poor accuracy of text classification

Images