摘要
文章提出了基于卷积神经网络的群众留言分类算法。首先,通过EDA技术进行数据增强;其次,用jieba和TF-IDF算法提取关键词;最后,通过embedding层、卷积层、池化层和全连接层实现对群众留言的多分类。实验结果表明,基于卷积神经网络的群众留言分类具有较好的分类效果。
This paper presents a classification algorithm of mass message based on convolution neural network.first,data enhancement is carried out by EDA technology;secondly,keywords are extracted by jieba and TF-IDF algorithms;finally,multi-classification of mass messages is realized by embedding layer,convolution layer,pool layer and full connection layer.The experimental results show that the classification of mass messages based on convolution neural network has better classification effect.
作者
代耀彬
朱燕燕
黄双华
Dai Yaobin;Zhu Yanyan;Huang Shuanghua(Hohai University,Nanjing 210098,China)
出处
《无线互联科技》
2020年第12期21-22,共2页
Wireless Internet Technology
关键词
群众留言分类
卷积神经网络
文本增强
public message classification
convolutional neural network
text enhancement
