短文本分类模型的优化及应用

被引：7
作者：
雷明珠
邵新慧
机构：
[1] 东北大学理学院
来源：
计算机应用研究 | 2021年 / 38卷 / 06期
关键词：
短文本; 神经主题模型; 特征融合;
D O I：
10.19734/j.issn.1001-3695.2020.06.0169
中图分类号：
TP391.1 [文字信息处理];
学科分类号：
081203 ; 0835 ;
摘要：
不同于长文本,短文本信息量缺乏,在研究中通常难以获得丰富的语义特征并且难以提取完整的句法特征,因此短文本分类模型的分类效果有待提升。针对这个问题进行了研究,基于Res LCNN模型进行改进,引入神经主题模型,并融合多个神经网络输出特征进行分类。首先,通过神经主题模型提取主题来丰富短文本的信息;其次,将主题信息储存在记忆网络中,并与序列信息进行融合,丰富文本的表示;最后,将其输入具有残差结构的卷积神经网络以及双向GRU中,提取局部以及全局的语义特征,在特征融合之后进行分类。该模型在Google网页搜索公开数据集中取得了较高的准确率和F1值,表明了改进模型在短文本分类任务中的有效性。
引用
收藏
页码：1775 / 1779
页数：5
相关论文
共 18 条
[1]  
Supervised Machine Learning: A Review of Classification Techniques[J] . Sotiris B. Kotsiantis. nformatica (Slovenia) . 2007 (3)
[2]  
Deep pyramid convolutional neural networks for text categorization . JOHNSON R,ZHANG T. Proceedings of the55th Annual Meeting of the Association for Computational Linguistics . 2017
[3]  
Hierarchical attention networks for document classification . Yang Z,Yang D,Dyer C. Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics:Human Language Technologies . 2016
[4]  
Character aware neural language models . Y.Kim,Y.Jernite,D.Sontag,A.M.Rush. Proceedings of the Thirtieth AAAI Conference on Artificial Intelligence . 2016
[5]  
Bag of Tricks for Efficient Text Classification . JOULIN A,GRAVE E,BOJANOWSKI P. Conference of the European Chapter of the Association for Computational Linguistics . 2017
[6]  
Investigating capsule networks with dynamic routing for text classification . Zhao Wei,Ye Jianbo,Yang Min. Proc of the 2018 Conference on Empirical Methods in Natural Language Processing . 2018
[7]  
Recurrent Convolutional Neural Networks for Text Classification . Siwei Lai,Liheng Xu,Kang Liu,Jun Zhao. 29th AAAI Conference on Artificial Intelligence . 2015
[8]  
A convolutional neural network for modelling sentences . Kalchbrenner N,Grefenstette E,Blunsom P. Proc of the 52nd Annual Meeting of the Association for Computational . 2014
[9]  
Convolutional neural networks for sentence classification . Kim Y. Conference on Empirical Methods in Natural Language Processing . 2014
[10]  
Combining naive Bayes and n-gram language models for text classification . Peng F,Schuurmans D. European Conference on Information Retrieval . 2003
12