Please wait a minute...
Data Analysis and Knowledge Discovery  2022, Vol. 6 Issue (2/3): 364-375    DOI: 10.11925/infotech.2096-3467.2021.0945
Current Issue | Archive | Adv Search |
End-to-End Aspect-Level Sentiment Analysis for E-Government Applications Based on BRNN
Shang Rongxuan1(),Zhang Bin2,Mi Jianing1
1School of Economics and Management, Harbin Institute of Technology, Harbin 150001, China
2Schoolof Public Administration and Law, Hunan Agricultural University, Changsha 410128, China
Download: PDF (1318 KB)   HTML ( 13 ) 
Export: BibTeX | EndNote (RIS)      
Abstract  

[Objective] This paper proposes an end-to-end aspect-level sentiment analysis method based on BRNN, aiming to conduct fine-grained sentiment analysis for reviews of government APPs. [Methods] First, we built a neural network containing a two-layer BRNN structure and three functional modules. Then, we recognized the boundary and sentiment tendency of the government APP reviews, as well as extracted aspect entities. [Results] The proposed E2E-ALSA model had excellent classification and generalization ability. Its precision, recall and F1-score all exceeded 0.93. [Limitations] The model can only jointly extract explicit aspect entities, while the implicit aspect extraction needs to be performed independently. The sample size needs to be expanded. [Conclusions] The proposed method could identify the users’ emotional needs and reactions to the e-government systems.
Key words： BRNN      End to End      Aspect-Level Sentiment Analysis      E-Government Application     
Received: 31 August 2021      Published: 14 April 2022
ZTFLH:  D630  
Fund:National Social Science Fund of China(17ZDA030)
Corresponding Authors: Shang Rongxuan,ORCID：0000-0002-3914-2650     E-mail: 564047413@qq.com
Service
E-mail this article
Add to my bookshelf
Add to citation manager
E-mail Alert
RSS
Articles by authors
Rongxuan Shang
Bin Zhang
Jianing Mi

Cite this article:

Shang Rongxuan, Zhang Bin, Mi Jianing. End-to-End Aspect-Level Sentiment Analysis for E-Government Applications Based on BRNN. Data Analysis and Knowledge Discovery, 2022, 6(2/3): 364-375.

URL:

https://manu44.magtech.com.cn/Jwk_infotech_wk3/EN/10.11925/infotech.2096-3467.2021.0945     OR     https://manu44.magtech.com.cn/Jwk_infotech_wk3/EN/Y2022/V6/I2/3/364

Comparison of ALSA Methods
Diagram of the Pipeline ALSA
Diagram of E2E-ALSA
Diagram of BRNN-based E2E-ALSA
输入 获得 自己 信息 很 方便 , 但是 登录 太卡 了
标签 O O S-POS O S-POS O O B-NEG E-NEG O
Example of Comment Text Labeling
RNN Structure
BRNN Structure
E2E-ALSA Based on BRNN Structure
比较项 结果
原评论 功能很全面,从方便百姓的角度出发很好！只是界面不是很好看,很多显示不全的地方。功能便利,界面不美观
分词后 '功能', '很', '全面', ',', '从', '方便', '百姓', '的', '角度', '出发', '很', '好', '！', '只是', '界面', '不是', '很', '好看', ',', '很多', '显示', '不全', '的', '地方', '。', '功能', '便利', ',', '界面', '不美观'
方面实体 功能 界面 功能 界面
实体位置 [1, 1] [15, 15] [26, 26] [29, 29]
情感极性 1 -1 1 -1
原评论 强制下载有意思吗?功能和下载率不匹配。也不能说完全没用,至少和便利无关。
分词后 '强制', '下载', '有意思', '吗', '?', '功能', '和', '下载', '率', '不', '匹配', '。', '也', '不能', '说', '完全', '没用', ',', '至少', '和', '便利', '无关', '。'
方面实体 强制下载 功能 下载率 便利
实体位置 [1, 2] [6, 6] [8, 8] [21, 21]
情感极性 -1 -1 -1 0
Example of Comment Text Labeling
w FT PS RS FS
1 0.986 7 0.763 4 0.766 8 0.765 1
3 0.988 9 0.927 9 0.923 8 0.925 8
5 0.995 5 0.953 9 0.937 2 0.945 5
7 0.991 1 0.950 0 0.928 3 0.936 7
9 0.993 5 0.953 7 0.923 8 0.938 4
The Variation of Network Performance with w
ε FT PS RS FS
0.3 0.991 5 0.913 6 0.901 3 0.907 4
0.4 0.992 4 0.934 3 0.892 4 0.912 8
0.5 0.995 5 0.953 9 0.937 2 0.945 5
0.6 0.995 9 0.941 2 0.932 7 0.936 9
0.7 0.993 5 0.934 3 0.892 4 0.912 8
The Variation of Network Performance with ε
参数 取值 参数 取值
BRNNS隐层神经元数 100 Drop-Out概率 0.35
BRNNT隐层神经元数 150 学习率 0.05
词向量长度 300 学习率衰减 0.01
输入窗长 5 梯度下降方法 随机梯度下降
ε 0.5 迭代次数 300
Network Training Parameters
方法模型 PS RS FS
HAST-TNet 0.885 2 0.853 6 0.869 1
LSTM-CRF 0.918 2 0.843 5 0.879 3
E2E-ALSA-LSTM 0.953 9 0.937 2 0.945 5
E2E-ALSA-GRU 0.934 8 0.898 6 0.916 3
Comparison of Experiments
Results of Partial Implicit Aspect Sentiment Analysis
[1] 赵玉林, 任莹, 周悦. 指尖上的形式主义:压力型体制下的基层数字治理——基于30个案例的经验分析[J]. 电子政务, 2020(3):100-109.
[1] ( Zhao Yulin, Ren Ying, Zhou Yue. Formalism at the Fingertips:Grassroots Digital Governance in a Pressure-Based System — An Empirical Analysis Based on 30 Cases[J]. E-Government, 2020(3):100-109.)
[2] 刘承宇, 冉琪. 电商客服端评论话语中的态度资源及其对潜在消费者消费决策的影响[J]. 北京科技大学学报(社会科学版), 2017, 33(6):1-7.
[2] ( Liu Chengyu, Ran Qi. Attitudinal Resources in Customers’ Remarks in CSC and Their Impacts on Potential Customers’ Purchase Decisions[J]. Journal of University of Science and Technology Beijing (Social Sciences Edition), 2017, 33(6):1-7.)
[3] di Sorbo A, Grano G, Aaron Visaggio C, et al. Investigating the Criticality of User-Reported Issues Through Their Relations with App Rating[J]. Journal of Software: Evolution and Process, 2021, 33(3):126-145.
[4] Zhao J, Liu K, Xu L H. Sentiment Analysis: Mining Opinions, Sentiments, and Emotions[J]. Computational Linguistics, 2016, 42(3):595-598.
doi: 10.1162/COLI_r_00259
[5] 张严, 李天瑞. 面向评论的方面级情感分析综述[J]. 计算机科学, 2020, 47(6):194-200.
[5] ( Zhang Yan, Li Tianrui. Review of Comment-Oriented Aspect-Based Sentiment Analysis[J]. Computer Science, 2020, 47(6):194-200.)
[6] Fu X H, Wei Y Z, Xu F, et al. Semi-Supervised Aspect-Level Sentiment Classification Model Based on Variational Autoencoder[J]. Knowledge-Based Systems, 2019, 171:81-92.
doi: 10.1016/j.knosys.2019.02.008
[7] 范昊, 李鹏飞. 基于FastText字向量与双向GRU循环神经网络的短文本情感分析研究——以微博评论文本为例[J]. 情报科学, 2021, 39(4):15-22.
[7] ( Fan Hao, Li Pengfei. Sentiment Analysis of Short Text Based on FastText Word Vector and Bidirectional GRU Recurrent Neural Network—Take the Microblog Comment Text as an Example[J]. Information Science, 2021, 39(4):15-22.)
[8] 陈苹, 冯林. 情感分析中的方面提取综述[J]. 计算机应用, 2018, 38(S2):84-88.
[8] ( Chen Ping, Feng Lin. Review of Aspect Extraction in Sentiment Analysis[J]. Journal of Computer Applications, 2018, 38(S2):84-88.)
[9] Tubishat M, Idris N, Abushariah M A M. Implicit Aspect Extraction in Sentiment Analysis: Review, Taxonomy, Oppportunities, and Open Challenges[J]. Information Processing & Management, 2018, 54(4):545-563.
doi: 10.1016/j.ipm.2018.03.008
[10] Huang S, Liu X L, Peng X P, et al. Fine-Grained Product Features Extraction and Categorization in Reviews Opinion Mining[C]// Proceedings of the 12th International Conference on Data Mining Workshops. IEEE, 2012: 680-686.
[11] Jin W, Ho H H. A Novel Lexicalized HMM-Based Learning Framework for Web Opinion Mining[C]//Proceedings of the 26th Annual International Conference on Machine Learning. New York: ACM Press, 2009: 465-472.
[12] Zheng S C, Wang F, Bao H Y, et al. Joint Extraction of Entities and Relations Based on a Novel Tagging Scheme[C]// Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics. 2017:1227-1236.
[13] 徐冰, 赵铁军, 王山雨, 等. 基于浅层句法特征的评价对象抽取研究[J]. 自动化学报, 2011, 37(10):1241-1247.
[13] ( Xu Bing, Zhao Tiejun, Wang Shanyu, et al. Extraction of Opinion Targets Based on Shallow Parsing Features[J]. Acta Automatica Sinica, 2011, 37(10):1241-1247.)
[14] Chen Z Y, Liu B. Mining Topics in Documents: Standing on the Shoulders of Big Data[C]// Proceedings of the 20th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2014: 1116-1125.
[15] Xu H, Liu B, Shu L, et al. Double Embeddings and CNN-Based Sequence Labeling for Aspect Extraction[C]// Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics. 2018: 592-598.
[16] Liu P F, Joty S, Meng H. Fine-Grained Opinion Mining with Recurrent Neural Networks and Word Embeddings[C]// Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing. 2015: 1433-1443.
[17] Wang Y Q, Huang M L, Zhu X Y, et al. Attention-Based LSTM for Aspect-Level Sentiment Classification[C]// Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing. 2016: 606-615.
[18] 唐慧丰, 谭松波, 程学旗. 基于监督学习的中文情感分类技术比较研究[J]. 中文信息学报, 2007, 21(6):88-94.
[18] ( Tang Huifeng, Tan Songbo, Cheng Xueqi. Research on Sentiment Classification of Chinese Reviews Based on Supervised Machine Learning Techniques[J]. Journal of Chinese Information Processing, 2007, 21(6):88-94.)
[19] 王婷, 杨文忠. 文本情感分析方法研究综述[J]. 计算机工程与应用, 2021, 57(12):11-24.
[19] ( Wang Ting, Yang Wenzhong. Review of Text Sentiment Analysis Methods[J]. Computer Engineering and Applications, 2021, 57(12):11-24.)
[20] Chen G M, Tian Y H, Song Y. Joint Aspect Extraction and Sentiment Analysis with Directional Graph Convolutional Networks[C]// Proceedings of the 28th International Conference on Computational Linguistics. 2020: 272-279.
[21] Schmitt M, Steinheber S, Schreiber K, et al. Joint Aspect and Polarity Classification for Aspect-Based Sentiment Analysis with End-to-End Neural Networks[C]// Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing. 2018.
[22] Schouten K, Frasincar F. Survey on Aspect-Level Sentiment Analysis[J]. IEEE Transactions on Knowledge and Data Engineering, 2016, 28(3):813-830.
doi: 10.1109/TKDE.2015.2485209
[23] Yan H, Dai J Q, Ji T, et al. A Unified Generative Framework for Aspect-Based Sentiment Analysis[C]// Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing. 2021: 2416-2429.
[24] Li X, Bing L D, Zhang W X, et al. Exploiting BERT for End-to-End Aspect-Based Sentiment Analysis[C]// Proceedings of the 5th Workshop on Noisy User-generated Text. 2019: 34-41.
[25] Devlin J, Chang M W, Lee K, et al. BERT: Pre-Training of Deep Bidirectional Transformers for Language Understanding[OL]. arXiv Preprint, arXiv:1810.04805.
[26] Liang Y L, Meng F D, Zhang J C, et al. A Dependency Syntactic Knowledge Augmented Interactive Architecture for End-to-End Aspect-Based Sentiment Analysis[J]. Neurocomputing, 2021, 454:291-302.
doi: 10.1016/j.neucom.2021.05.028
[27] 周安桥. 基于深度学习的属性级情感分析[D]. 大连: 大连理工大学, 2020.
[27] ( Zhou Anqiao. Aspect-Level Sentiment Analysis Based on Deep Learning[D]. Dalian: Dalian University of Technology, 2020.)
[28] Ma X, Hovy E. End-to-End Sequence Labeling via Bi-Directional LSTM-CNNS-CRF[C]// Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics. 2016:1064-1074.
[29] Li S, Zhao Z, Hu R F, et al. Analogical Reasoning on Chinese Morphological and Semantic Relations[C]// Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics. 2018:138-143.
[30] Li S, Zhao Z, Hu R F, et al. Analogical Reasoning on Chinese Morphological and Semantic Relations[C]// Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics. 2018: 138-143.
[31] Qiu Y Y, Li H Z, Li S, et al. Revisiting Correlations Between Intrinsic and Extrinsic Evaluations of Word Embeddings[C]// Proceedings of the 17th China National Conference and 6th International Symposium. 2018: 209-221.
[32] Li X, Bing L, Li P, et al. Aspect Term Extraction with History Attention and Selective Transformation[C]// Proceedings of the 27th International Joint Conference on Artificial Intelligence. 2018: 4149-4200.
[33] Tong J N, Chen W, Wei Z H. Attentional Transformer Networks for Target-Oriented Sentiment Classification[C]// Proceedings of the 7th CCF Conference on Big Data. 2019: 271-284.
[34] Davis V F D. A Theoretical Extension of the Technology Acceptance Model: Four Longitudinal Field Studies[J]. Management Science, 2000, 46(2):186-204.
doi: 10.1287/mnsc.46.2.186.11926
[1] Yu Bengong, Zhang Shuwen. Aspect-Level Sentiment Analysis Based on BAGCNN[J]. 数据分析与知识发现, 2021, 5(12): 37-47.
  Copyright © 2016 Data Analysis and Knowledge Discovery   Tel/Fax:(010)82626611-6626，82624938   E-mail:jishu@mail.las.ac.cn