Convolution-based Memory Network for Aspect-based Sentiment Analysis

被引:49
作者
Fan, Chuang [1 ]
Gao, Qinghong [1 ]
Du, Jiachen [1 ]
Gui, Lin [2 ]
Xu, Ruifeng [1 ]
Wong, Kam-Fai [3 ]
机构
[1] Harbin Inst Technol, Shenzhen Grad Sch, Harbin, Heilongjiang, Peoples R China
[2] Aston Univ, Birmingham, W Midlands, England
[3] Chinese Univ Hong Kong, Shatin, Hong Kong, Peoples R China
来源
ACM/SIGIR PROCEEDINGS 2018 | 2018年
基金
欧盟地平线“2020”; “创新英国”项目; 中国国家自然科学基金;
关键词
sentiment analysis; memory network; convolutional operation;
D O I
10.1145/3209978.3210115
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Memory networks have shown expressive performance on aspect based sentiment analysis. However, ordinary memory networks only capture word-level information and lack the capacity for modeling complicated expressions which consist of multiple words. Targeting this problem, we propose a novel convolutional memory network which incorporates an attention mechanism. This model sequentially computes the weights of multiple memory units corresponding to multi-words. This model may capture both words and multi-words expressions in sentences for aspect-based sentiment analysis. Experimental results show that the proposed model outperforms the state-of-the-art baselines.
引用
收藏
页码:1161 / 1164
页数:4
相关论文
共 12 条
[1]  
[Anonymous], ABS151201100 CORR
[2]  
[Anonymous], 2015, ADV NEURAL INFORM PR
[3]  
Chen P., 2017, P 2017 C EMP METH NA, P452, DOI [10.18653/v1/D17-1047, DOI 10.18653/V1/D17-1047]
[4]  
Dong L, 2014, PROCEEDINGS OF THE 52ND ANNUAL MEETING OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS, VOL 2, P49
[5]  
Gui L., 2017, ABS170805482 CORR, P1593
[6]  
Hu M, 2004, P 10 ACM SIGKDD INT, P168
[7]  
Kim Y., 2014, P C EMP METH NAT LAN, P1746, DOI DOI 10.3115/V1/D14-1181
[8]  
Kiritchenko S, 2014, P 8 INT WORKSH SEM E, P437, DOI DOI 10.3115/V1/S14-2076
[9]   Deep Memory Networks for Attitude Identification [J].
Li, Cheng ;
Guo, Xiaoxiao ;
Mei, Qiaozhu .
WSDM'17: PROCEEDINGS OF THE TENTH ACM INTERNATIONAL CONFERENCE ON WEB SEARCH AND DATA MINING, 2017, :671-680
[10]  
Pennington J., 2014, 2014 C EMP METH NAT, P43