Text classification method, apparatus, computer device, and storage medium

Abstract

The embodiment of the invention discloses a text classification method, a device, a computer device and a storage medium. The method comprises the following steps: the input text is processed by wordsegmentation and part-of-speech tagging to obtain a word segmentation list, wherein the word segmentation list comprises a word obtained by word segmentation of the input text and a part-of-speech ofthe word; Obtaining a word vector of each word and a word vector of a part of speech of each word in the word segmentation list; Obtaining a word vector matrix composed of a splicing word vector of each word in the word segmentation list, wherein, the splicing word vector is obtained by splicing a word vector of a word and a word vector of a part of speech of the word; Inputting the word vector matrix to Bi-LSTM to obtain a text feature vector of each word in the word segmentation list; Obtaining a text classification result of the input text according to a text feature vector of each word inthe word segmentation list. By implementing the method of the embodiment of the invention, the accuracy of the text classification can be improved.

Description

technical field [0001] The present invention relates to the technical field of artificial intelligence, in particular to a text classification method, device, computer equipment and storage medium. Background technique [0002] At present, the neural network models commonly used in text classification include CNN+Word-Embedding, RNN+Word-Embedding, MLP+Word-Embedding and other structures. The usual way to use this type of structure is to convert and map the text into a real number space after word segmentation, and then combine it into a floating-point matrix or vector that can be accepted by the neural network as an input, and then obtain the probability density distribution of the classification through the calculation of the neural network model , use gradient descent or some improved training methods to optimize the model until convergence during training. [0003] At present, the neural network + Word-Embedding model commonly used in deep learning mainly has the follow...

AI Technical Summary

Problems solved by technology

[0003] At present, the neural network + Word-Embedding model commonly used in deep learning mainly has the following problems in text classification. The MLP full-link neural network completely ignores the sequence of word vectors when extracting text features, and treats the entire text as only one The collection of vocabulary, which will lead to the inability to obtain the features in many semantic expressions, which will affect the classification accuracy

Images