Text classification method, system, readable storage medium and electronic device

Abstract

A text classification method, system, readable storage medium, and electronic equipment include: taking a phrase structure tree corresponding to a sentence text, wherein the sentence text includes one or more tags, each tag corresponds to a word vector, and each word vector Used as the input of the leaf node in the phrase structure tree; the first operation process is performed on the leaf node in the phrase structure tree by a mark encoder, so that the word vector of the sentence text is converted into a corresponding sentence vector; The internal node in the phrase structure tree executes the second operation process to obtain the output of the root node in the phrase structure tree according to the sentence vector; the classifier is trained according to the output of the root node. The invention can improve the accuracy rate of text classification.

Description

technical field [0001] The invention relates to the field of data processing, in particular to a neural network-based text classification method, system, readable storage medium and electronic equipment. Background technique [0002] Most neural networks used for sentence representation usually fall into one of the following categories: sequential models, convolutional models, and recurrent models. Recently, sequence models have witnessed widespread applications in Natural Language Processing (NLP), such as text classification, machine translation, question answering, etc. In these methods, a Recurrent Neural Network (RNN) takes each word as input, aggregates the word with its previous state, and finally outputs its result for an entire sentence. The synthesized results as fixed-length vectors contain rich semantic information and are used in subsequent NLP tasks. [0003] However, in natural language understanding, recurrent neural networks including LSTMs all process wor...

AI Technical Summary

Problems solved by technology

Obviously, the existing models do not make full use of the sentence structure, or are not suitable for unrestricted phrase structure trees

Images