Construction of Text Emotion Classification Model Based on Convolutional Neural Network

Abstract

Text emotion analysis transforms a text sequence of indefinite length into text category which is one of the key research problems in the field of natural language processing. With the wide application of deep learning technology in natural language processing the text emotion analysis model based on deep learning has made a new breakthrough. This paper builds a basic framework of text emotion analysis and describes it from two aspects data preprocessing and design of network structure of revolutionary neural network. Data preprocessing mainly involves word segmentation model and word embedding. Design of network structure of revolutionary network is the basic structure of the revolutionary network (CNN) including input layer convolution layer pool layer full connection layer and output layer. Finally the feasibility of the method is verified by experiments.