Text classification model training method and device, electronic equipment and storage medium

Abstract

The invention discloses a text classification model training method and device, electronic equipment and a storage medium, and belongs to the technical field of model training. The text classification model training method comprises the steps that a first text characterization network, a second text characterization network and a third text characterization network are constructed, the first text characterization network and the second text characterization network are characterization networks with similar semantics, and the second text characterization network and the third text characterization network are characterization networks with different semantics; and inputting the text data set into the first text representation network, the second text representation network and the third text representation network to train the text classification model so as to obtain a trained text classification model.

Description

technical field [0001] The present application belongs to the technical field of model training, and specifically relates to a training method, device, electronic device and storage medium for a text classification model. Background technique [0002] Text classification is one of the most basic and important tasks in natural language processing. After training a model through machine learning or deep learning, the trained model can label text units such as phrases, sentences, paragraphs, and even articles. [0003] In the prior art, feature representations with similar semantics are constructed in the process of training the model. Only the samples with similar semantics constructed by the feature representation network with the same throw value are considered, and the samples with different semantics cannot be trained. In the learning process, the samples with different semantics cannot be fully trained and learned, so that the accuracy of the text classification model obt...

AI Technical Summary

Problems solved by technology

[0003] In the existing technology, feature representations with similar semantics are constructed in the process of training the model. Only considering the samples with similar semantics constructed by the feature representation network with the same thrown value, it is impossible to train samples with different semantics, resulting in the failure of the text classification model. In the learning process, samples with different semantics cannot be fully trained and learned, which makes the accuracy of the trained text classification model poor

Images