Chinese text emotion classification method based on multi-kernel double-layer convolutional neural network

Abstract

The invention discloses a Chinese text emotion classification method based on a multi-kernel double-layer convolutional neural network, and the method comprises the steps of obtaining a Chinese text data set, and carrying out the preprocessing of the Chinese text data set; performing feature extraction on the preprocessed Chinese text data set through a plurality of different feature extraction methods to obtain a plurality of different word vector matrixes; constructing a Chinese text emotion classifier based on a multi-kernel double-layer convolutional neural network, and inputting the extracted multiple different word vector matrixes into the Chinese text emotion classifier to complete training of the Chinese text emotion classifier, wherein the trained Chinese text emotion classifier is used for carrying out emotion classification on the Chinese text. According to the invention, the significant features of the Chinese text can be reserved, the feature dimension is reduced, the problems that feature extraction is too abstract and information is lost are avoided, the situation that feature information extraction is insufficient is solved, and rapid and accurate classification of Chinese text emotions is achieved.

Description

technical field [0001] The invention relates to the technical field of Chinese text emotion classification, in particular to a Chinese text emotion classification method based on a multi-core double-layer convolutional neural network. Background technique [0002] With the rapid development of the Internet, especially the advent of the Web 2.0 era, network information dissemination has developed from one-way information release to dynamic information interaction. Users are no longer just readers of network content, but also producers of network content. Network communication platforms such as forums, Weibo, WeChat, and e-commerce reviews continue to emerge, and people are becoming more and more accustomed to posting subjective remarks on the Internet to express their views on events and policies they are concerned about or the goods and services they purchase. and views. Data rich in emotional information generated by a large number of users on the Internet provides new opp...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a Chinese text emotion classification method based on a multi-core double-layer convolutional neural network, to solve the problems of the prior art, to retain the salient features of the Chinese text, reduce the feature dimension, and avoid feature extraction that is too abstract and The problem of information loss has also solved the problem of insufficient feature information extraction through a variety of different convolution kernels. It has the advantages of good processing of discrete data, latent semantic analysis, consideration of word polysemy, and suitable for processing large-scale data. Fast and Accurate Classification of Sentiment in Chinese Text

Images