摘要
文本情感分析是自然语言处理中的一项重要任务,而循环神经网络和卷积神经网络是自然语言处理中常用的两种深度学习模型。本文提出了一种残差网络、多层双向门控递归单元和文本卷积神经网络相结合的残差图卷积神经网络,并在多个英文、中文数据集上获得了良好的分类性能。
Text emotion analysis is an important task in natural language processing,and recurrent neural network and convolutional neural network are two deep learning models commonly used in natural language processing.In this paper,a residual graph convolution neural network combining residual network,multilayer bidirectional gated recursive unit and text convolution neural network is proposed.It has achieved good classification performance on many English and Chinese data sets.
作者
覃光明
QIN Guangming(Department of Architectural Engineering,Guangxi Modern Vocational and Technical College,Hechi,China,547000)
出处
《福建电脑》
2022年第2期20-24,共5页
Journal of Fujian Computer
基金
广西职业教育教学改革研究项目(No.GXGZJG2021B118)资助。
关键词
情感分析
神经网络
自然语言处理
卷积神经网络
Emotional Analysis
Neural Network
Natural Language Processing
Convolutional Neural Network
