Sentence-Level Sentiment Analysis via BERT and BiGRU

被引:6
作者
Shen, Jianghong [1 ]
Liao, Xiaodong
Tao, Zhuang
机构
[1] Fujian Normal Univ, Coll Photon & Elect Engn, Minist Educ, Key Lab Optoelect Sci & Technol Med, Fuzhou 350007, Peoples R China
来源
2019 INTERNATIONAL CONFERENCE ON IMAGE AND VIDEO PROCESSING, AND ARTIFICIAL INTELLIGENCE | 2019年 / 11321卷
关键词
sentiment analysis; contextualized embedding; BERT; BiGRU;
D O I
10.1117/12.2550215
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Sentiment analysis is a significant task in nature language processing (NLP). Acquiring high quality word representations is a key point in the task. Specially we find that the same word has different meaning in different sentence, which should be recognized by computer. This idea cannot be done well by traditional way of word embeddings. In this paper, we propose a BERT(Bidirectional Encoder Representation from Transformers) + BiGRU (Bidirectional Gated Recurrent Unit) model which first put words into vector via BERT model, from which we can gain the contextualized embeddings, then perform the sentiment analysis by BiGRU. Experimental results prove that compared with various of different methods, our model has the best performing.
引用
收藏
页数:6
相关论文
共 19 条
[1]  
[Anonymous], COMPUTER SCI PAPERS
[2]  
[Anonymous], P EACL
[3]  
[Anonymous], P SEMEVAL 2016
[4]  
[Anonymous], 2017, CORR
[5]  
[Anonymous], P CONLL
[6]  
[Anonymous], 2015, P NIPS
[7]  
[Anonymous], 2012, SENTIMENT ANAL OPINI, DOI DOI 10.2200/S00416ED1V01Y201204HLT016
[8]  
[Anonymous], 2017, P 15 C EUR ASS COMP
[9]  
[Anonymous], CORR
[10]  
[Anonymous], 2014, P 2014 C EMP METH NA