Text sentiment classification method and system

Abstract

The invention discloses a text sentiment classification method and system, and the method comprises the steps: dividing a text sentence in terms of words, and mapping each word into a word vector; extracting keywords in the text sentences, respectively constructing a word vector attention matrix, a position attention matrix and a part-of-speech attention matrix according to word vectors of the keywords, positions of the keywords in the text sentences and emotion part-of-speech types to which the keywords belong, and fusing the word vector attention matrix, the position attention matrix and thepart-of-speech attention matrix to construct a first feature; adopting a BiGRU network to obtain a second feature according to the context semantic information of the keyword; and classifying the emotion types of the text sentences to be tested by adopting a multi-attention convolutional neural network model trained by taking the first features and the second features as a training set. Accordingto the method, a keyword sentiment classification first-dimensional feature is obtained in combination with a CNN model of a multi-attention mechanism, an initial sentence sentiment classification second-dimensional feature is obtained through BiGRU, the two dimensional features are fused, the text deep-level semantic perception capability is improved, and then the text sentiment classification accuracy is improved.

Description

technical field [0001] The present invention relates to the technical fields of natural language processing and deep learning, in particular to a text emotion classification method and system. Background technique [0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art. [0003] With the continuous development of social networks, the role of Internet users has also changed from the original information recipients to information creators. Internet users are accustomed to expressing opinions through the Internet and publishing content with emotional attitudes on Internet platforms, forming a short text-based The main expression method, in which the amount of text information data is large, and the arrangement of text content is scattered and messy, it is difficult to distinguish and organize manually. Therefore, how to use natural language processing related technologies to anal...

AI Technical Summary

Problems solved by technology

The variant BiGRU model of LSTM can effectively extract the semantic information before and after the text, and has a stronger perception of the overall semantics of the sentence, and the model is more streamlined than the LSTM; although the BiGRU model can effectively learn the information closely related to the emotional tendency Text sequence semantic features, but cannot further learn the local semantic features of the text

Images