基于MC_E_ImprovDPC_B_SV模型的中文情感分类方法  

Chinese emoticn classfication method base on MC_E_ImprovDPC_B_SV model

作　　者：朱海东 郑虹[1] 侯秀萍[1] ZHU Haidong;ZHENG Hong;HOU Xiuping(School of Computer Science & Engineering, Changchun University of Technology, Changchun 130012, China)

机构地区：[1]长春工业大学计算机科学与工程学院,吉林长春130012

出　　处：《长春工业大学学报》2022年第1期72-80,共9页Journal of Changchun University of Technology

基　　金：吉林省教育厅项目(JJKH20200677KJ)。

摘　　要：为解决CNN和RNN无法同时兼顾局部和全局特征提取、传统情感分类模型不能充分利用原始评论文本中所蕴含的情感和语义信息、字词多义性调整、短文本语义稀疏等问题。提出一种基于ERNIE预训练模型和改进DPCNN的多通道的中文情感分类模型(MC_E_ImprovDPC_B_SV)来提高分类准确率,并实现准确分类。In order to solve the problems that CNN and RNN cannot simultaneously take into account the local and global feature extraction,the traditional emotion classification model cannot make full use of the emotional and semantic information contained in the original comment text,the polysemy adjustment of words,and the semantic sparseness of short texts.Therefore,a multi-channel Chinese sentiment classification model based on ERNIE pre-training model and improved DPCNN(MC_E_ImprovDPC_B_SV)is proposed to improve the classification accuracy and achieve accurate classification.

关 键 词：中文情感分类 多通道提取特征 改进的DPCNN 语义特征 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]