Journal of Computer Applications ›› 2024, Vol. 44 ›› Issue (6): 1786-1795.DOI: 10.11772/j.issn.1001-9081.2023050638
Special Issue: 人工智能
• Artificial intelligence • Previous Articles Next Articles
Tianci KE1,2, Jianhua LIU1,2(), Shuihua SUN1,2, Zhixiong ZHENG1,2, Zijie CAI1,2
Received:
2023-05-24
Revised:
2023-08-14
Accepted:
2023-08-21
Online:
2023-08-25
Published:
2024-06-10
Contact:
Jianhua LIU
About author:
KE Tianci, born in 1999, M. S. candidate. His research interests include aspect-level sentiment analysis.Supported by:
柯添赐1,2, 刘建华1,2(), 孙水华1,2, 郑智雄1,2, 蔡子杰1,2
通讯作者:
刘建华
作者简介:
柯添赐(1999—),男,福建莆田人,硕士研究生,CCF会员,主要研究方向:方面级情感分析基金资助:
CLC Number:
Tianci KE, Jianhua LIU, Shuihua SUN, Zhixiong ZHENG, Zijie CAI. Aspect-level sentiment analysis model combining strong association dependency and concise syntax[J]. Journal of Computer Applications, 2024, 44(6): 1786-1795.
柯添赐, 刘建华, 孙水华, 郑智雄, 蔡子杰. 融合强关联依赖和简洁语法的方面级情感分析模型[J]. 《计算机应用》唯一官方网站, 2024, 44(6): 1786-1795.
Add to citation manager EndNote|Ris|BibTeX
URL: https://www.joca.cn/EN/10.11772/j.issn.1001-9081.2023050638
词性 | 词性解释 | 词性 | 词性解释 |
---|---|---|---|
JJ | 形容词 | VBN | 动词,过去分词 |
JJR | 形容词比较级 | VBP | 动词,非第三人称单数 |
JJS | 形容词最高级 | VB | 动词,原形 |
VBD | 动词,过去式 |
Tab. 1 POS list for sentiment words
词性 | 词性解释 | 词性 | 词性解释 |
---|---|---|---|
JJ | 形容词 | VBN | 动词,过去分词 |
JJR | 形容词比较级 | VBP | 动词,非第三人称单数 |
JJS | 形容词最高级 | VB | 动词,原形 |
VBD | 动词,过去式 |
类别 | 属性 | 属性解释 |
---|---|---|
依赖关系 | cop | 系动词 |
det | 决定词,如冠词等 | |
cc | 并列关系 | |
punct | 标点符号 | |
词性 | PRP$ | 物主代词 |
PRP | 人称代词 | |
WDT | 限定词 | |
WP | 疑问代词 | |
WRB | 疑问副词 | |
POS | 所有格结束词 | |
PDT | 前置限定词 | |
RP | 分词 |
Tab.2 Joint list of combining parts-of-speech and dependencies
类别 | 属性 | 属性解释 |
---|---|---|
依赖关系 | cop | 系动词 |
det | 决定词,如冠词等 | |
cc | 并列关系 | |
punct | 标点符号 | |
词性 | PRP$ | 物主代词 |
PRP | 人称代词 | |
WDT | 限定词 | |
WP | 疑问代词 | |
WRB | 疑问副词 | |
POS | 所有格结束词 | |
PDT | 前置限定词 | |
RP | 分词 |
数据集 | 不同句子类别样本数 | 不同标签类别样本数 | |||||
---|---|---|---|---|---|---|---|
多方面词 | 单方面词 | 总数 | 积极 | 消极 | 中立 | ||
Restaurant | 训练集 | 971 | 1 009 | 1 980 | 2 164 | 807 | 637 |
测试集 | 315 | 284 | 599 | 727 | 196 | 196 | |
Laptop | 训练集 | 538 | 916 | 1 454 | 937 | 851 | 455 |
测试集 | 150 | 259 | 409 | 337 | 128 | 167 | |
MAMS | 训练集 | 4 297 | 0 | 4 297 | 3 380 | 2 764 | 5 042 |
验证集 | 500 | 0 | 500 | 403 | 325 | 604 | |
测试集 | 500 | 0 | 500 | 400 | 329 | 607 | |
训练集 | 0 | 6 051 | 6 051 | 1 507 | 1 528 | 3 016 | |
测试集 | 0 | 677 | 677 | 172 | 169 | 336 |
Tab.3 Experimental data statistics
数据集 | 不同句子类别样本数 | 不同标签类别样本数 | |||||
---|---|---|---|---|---|---|---|
多方面词 | 单方面词 | 总数 | 积极 | 消极 | 中立 | ||
Restaurant | 训练集 | 971 | 1 009 | 1 980 | 2 164 | 807 | 637 |
测试集 | 315 | 284 | 599 | 727 | 196 | 196 | |
Laptop | 训练集 | 538 | 916 | 1 454 | 937 | 851 | 455 |
测试集 | 150 | 259 | 409 | 337 | 128 | 167 | |
MAMS | 训练集 | 4 297 | 0 | 4 297 | 3 380 | 2 764 | 5 042 |
验证集 | 500 | 0 | 500 | 403 | 325 | 604 | |
测试集 | 500 | 0 | 500 | 400 | 329 | 607 | |
训练集 | 0 | 6 051 | 6 051 | 1 507 | 1 528 | 3 016 | |
测试集 | 0 | 677 | 677 | 172 | 169 | 336 |
参数 | 值 | 参数 | 值 |
---|---|---|---|
词嵌入维度 | 768 | batch_size | 16 |
学习率 | 2×10-5 | GAT注意力头数 | 4 |
L2正则化权值 | 1×10-5 | 优化器 | Adamax |
dropout rate | 0.1 |
Tab. 4 Hyperparameters setting
参数 | 值 | 参数 | 值 |
---|---|---|---|
词嵌入维度 | 768 | batch_size | 16 |
学习率 | 2×10-5 | GAT注意力头数 | 4 |
L2正则化权值 | 1×10-5 | 优化器 | Adamax |
dropout rate | 0.1 |
名称 | 信息 | 名称 | 信息 |
---|---|---|---|
操作系统 | Windows 10 | 开发工具 | PyCharm |
显卡 | NVIDIA RTX 3070 | 深度学习框架 | PyTorch 1.10.0 |
显存 | 8.0 GB |
Tab. 5 Experimental environment
名称 | 信息 | 名称 | 信息 |
---|---|---|---|
操作系统 | Windows 10 | 开发工具 | PyCharm |
显卡 | NVIDIA RTX 3070 | 深度学习框架 | PyTorch 1.10.0 |
显存 | 8.0 GB |
类别 | 模型 | Restaurant | Laptop | MAMS | |||||
---|---|---|---|---|---|---|---|---|---|
Acc | MF1 | Acc | MF1 | Acc | MF1 | Acc | MF1 | ||
w/o Syn. | SVM | 80.16 | — | 70.49 | — | 63.40 | 63.30 | — | — |
ATAE-LSTM | 77.20 | — | 68.70 | — | — | — | — | — | |
TNet | 80.69 | 71.27 | 76.54 | 71.75 | 74.97 | 73.60 | — | — | |
AEN | 80.98 | 72.14 | 73.51 | 69.04 | 72.83 | 69.81 | 66.72 | — | |
AEN-BERT | 83.12 | 73.76 | 79.93 | 76.31 | 74.71 | 73.13 | 70.06 | — | |
BERT-SPC | 84.46 | 76.98 | 78.99 | 75.03 | 73.55 | 72.14 | 82.82 | 81.90 | |
w Syn. | PhraseRNN | 66.20 | 59.32 | — | — | — | — | — | — |
CDT | 82.30 | 74.02 | 77.19 | 72.99 | 74.66 | 73.66 | 80.70 | 79.79 | |
MHAGCN | 81.43 | 73.15 | 75.85 | 71.38 | 73.03 | 70.96 | — | — | |
MHAGCN-BERT | 82.57 | 75.83 | 79.06 | 75.70 | 74.53 | 73.75 | — | — | |
BiSyn-GAT+* | 86.24 | 79.87 | 79.91 | 75.82 | 75.92 | 74.98 | 83.46 | 83.13 | |
DMF-GAT-BERT | 86.10 | 80.17 | 80.38 | 77.20 | 76.22 | 75.10 | 83.86 | 83.19 | |
SADCS | 87.58 | 82.15 | 82.19 | 78.44 | 76.31 | 75.54 | 84.30 | 83.59 |
Tab. 6 Comparative experimental results of models of different categories on different datasets
类别 | 模型 | Restaurant | Laptop | MAMS | |||||
---|---|---|---|---|---|---|---|---|---|
Acc | MF1 | Acc | MF1 | Acc | MF1 | Acc | MF1 | ||
w/o Syn. | SVM | 80.16 | — | 70.49 | — | 63.40 | 63.30 | — | — |
ATAE-LSTM | 77.20 | — | 68.70 | — | — | — | — | — | |
TNet | 80.69 | 71.27 | 76.54 | 71.75 | 74.97 | 73.60 | — | — | |
AEN | 80.98 | 72.14 | 73.51 | 69.04 | 72.83 | 69.81 | 66.72 | — | |
AEN-BERT | 83.12 | 73.76 | 79.93 | 76.31 | 74.71 | 73.13 | 70.06 | — | |
BERT-SPC | 84.46 | 76.98 | 78.99 | 75.03 | 73.55 | 72.14 | 82.82 | 81.90 | |
w Syn. | PhraseRNN | 66.20 | 59.32 | — | — | — | — | — | — |
CDT | 82.30 | 74.02 | 77.19 | 72.99 | 74.66 | 73.66 | 80.70 | 79.79 | |
MHAGCN | 81.43 | 73.15 | 75.85 | 71.38 | 73.03 | 70.96 | — | — | |
MHAGCN-BERT | 82.57 | 75.83 | 79.06 | 75.70 | 74.53 | 73.75 | — | — | |
BiSyn-GAT+* | 86.24 | 79.87 | 79.91 | 75.82 | 75.92 | 74.98 | 83.46 | 83.13 | |
DMF-GAT-BERT | 86.10 | 80.17 | 80.38 | 77.20 | 76.22 | 75.10 | 83.86 | 83.19 | |
SADCS | 87.58 | 82.15 | 82.19 | 78.44 | 76.31 | 75.54 | 84.30 | 83.59 |
模型 | Restaurant | Laptop | ||
---|---|---|---|---|
Acc | MF1 | Acc | MF1 | |
w/o SC | 86.15 | 79.85 | 77.81 | 73.10 |
w/o GAT | 85.15 | 78.31 | 78.59 | 75.21 |
w/o BiLSTM | 85.34 | 79.22 | 79.22 | 75.99 |
w/o Cross-Att | 85.61 | 79.12 | 79.38 | 75.43 |
SADCS | 87.58 | 82.15 | 82.19 | 78.44 |
Tab.7 Ablation experimental results of SADCS
模型 | Restaurant | Laptop | ||
---|---|---|---|---|
Acc | MF1 | Acc | MF1 | |
w/o SC | 86.15 | 79.85 | 77.81 | 73.10 |
w/o GAT | 85.15 | 78.31 | 78.59 | 75.21 |
w/o BiLSTM | 85.34 | 79.22 | 79.22 | 75.99 |
w/o Cross-Att | 85.61 | 79.12 | 79.38 | 75.43 |
SADCS | 87.58 | 82.15 | 82.19 | 78.44 |
模型 | Restaurant | Laptop | MAMS | |||||
---|---|---|---|---|---|---|---|---|
Acc | MF1 | Acc | MF1 | Acc | MF1 | Acc | MF1 | |
w/o Single-Asp | 85.79 | 79.75 | 78.28 | 74.27 | 73.11 | 71.80 | 84.30 | 83.59 |
w/o Multi-Asp | 86.32 | 79.83 | 78.44 | 74.96 | 76.31 | 75.54 | 84.00 | 83.42 |
w/o Single-Asp & Multi-Asp | 86.14 | 79.97 | 78.28 | 75.11 | 73.11 | 71.80 | 84.00 | 83.42 |
w/o prune | 85.96 | 79.89 | 78.75 | 75.67 | 75.44 | 74.84 | 84.30 | 83.59 |
w/o Single-Asp & prune | 85.97 | 79.66 | 77.66 | 73.85 | 74.42 | 72.73 | 84.30 | 83.59 |
w/o Multi-Asp & prune | 86.77 | 80.80 | 76.88 | 72.76 | 75.44 | 74.84 | 84.00 | 83.42 |
SADCS | 87.58 | 82.15 | 82.19 | 78.44 | 76.31 | 75.54 | 84.30 | 83.59 |
Tab. 8 Ablation experimental results of strong association and concise syntax graph
模型 | Restaurant | Laptop | MAMS | |||||
---|---|---|---|---|---|---|---|---|
Acc | MF1 | Acc | MF1 | Acc | MF1 | Acc | MF1 | |
w/o Single-Asp | 85.79 | 79.75 | 78.28 | 74.27 | 73.11 | 71.80 | 84.30 | 83.59 |
w/o Multi-Asp | 86.32 | 79.83 | 78.44 | 74.96 | 76.31 | 75.54 | 84.00 | 83.42 |
w/o Single-Asp & Multi-Asp | 86.14 | 79.97 | 78.28 | 75.11 | 73.11 | 71.80 | 84.00 | 83.42 |
w/o prune | 85.96 | 79.89 | 78.75 | 75.67 | 75.44 | 74.84 | 84.30 | 83.59 |
w/o Single-Asp & prune | 85.97 | 79.66 | 77.66 | 73.85 | 74.42 | 72.73 | 84.30 | 83.59 |
w/o Multi-Asp & prune | 86.77 | 80.80 | 76.88 | 72.76 | 75.44 | 74.84 | 84.00 | 83.42 |
SADCS | 87.58 | 82.15 | 82.19 | 78.44 | 76.31 | 75.54 | 84.30 | 83.59 |
1 | 王璐,马宏伟,吕欢欢.方面级文本情感分析综述[J].计算机应用,2022,42(S2):1-9. |
WANG L, MA H W, LYU H H. Summary of aspect-based sentiment analysis[J]. Journal of Computer Applications, 2022, 42(S2): 1-9. | |
2 | 谭翠萍.文本细粒度情感分析研究综述[J].大学图书馆学报,2022,40(4):85-99,119. |
TAN C P. Review of fine-grained sentiment analysis based on text[J]. Journal of Academic Libraries, 2022, 40(4): 85-99,119. | |
3 | JIANG L, YU M, ZHOU M, et al. Target-dependent Twitter sentiment classification[C]// Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies. Stroudsberg: ACL, 2011: 151-160. |
4 | TANG D, QIN B, FENG X, et al. Effective LSTMs for target-dependent sentiment classification[C]// Proceedings of the 26th International Conference on Computational Linguistics: Technical Papers. [S.l.]: The COLING 2016 Organizing Committee, 2016: 3298-3307. |
5 | MA D, LI S, ZHANG X, et al. Interactive attention networks for aspect-level sentiment classification[C]// Proceedings of the 26th International Joint Conference on Artificial Intelligence. Palo Alto: AAAI Press, 2017: 4068-4074. |
6 | LI L, LIU Y, ZHOU A Q. Hierarchical attention based position-aware network for aspect-level sentiment analysis[C]// Proceedings of the 22nd Conference on Computational Natural Language Learning. Stroudsberg: ACL, 2018: 181-189. |
7 | LI J, WANG X, TU Z, et al. On the diversity of multi-head attention[J]. Neurocomputing. 2021, 454: 14-24. |
8 | CHEN Y, ZHUANG T, GUO K. Memory network with hierarchical multi-head attention for aspect-based sentiment analysis[J]. Applied Intelligence, 2021, 51(7): 4287-4304. |
9 | SHAO D, AN Q, HUANG K, et al. Aspect-level sentiment analysis for based on joint aspect and position hierarchy attention mechanism network[J]. Journal of Intelligent & Fuzzy Systems, 2022, 42(3): 2207-2218. |
10 | LIN T, SUN A, WANG Y. Aspect-based sentiment analysis through EDU-level attentions[C]// Advances in Knowledge Discovery and Data Mining: 26th Pacific-Asia Conference. Berlin: Springer, 2022: 156-168. |
11 | HUANG B, CARLEY K. Syntax-aware aspect level sentiment classification with graph attention networks[C]// Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing. Stroudsberg: ACL, 2019: 5469-5477. |
12 | YAN H, DAI J, 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 (Volume 1: Long Papers). Stroudsberg: ACL, 2021: 2416-2429. |
13 | MA L, RABBANY R, ROMERO-SORIANO A. Graph attention networks with positional embeddings [C]// Proceedings of the 2021 Pacific-Asia Conference on Knowledge Discovery and Data Mining. Cham: Springer, 2021: 514-527. |
14 | BAI X, LIU P, ZHANG Y. Investigating typed syntactic dependencies for targeted sentiment classification using graph attention neural network[J]. IEEE/ACM Transactions on Audio, Speech, and Language Processing, 2020, 29: 503-514. |
15 | VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need [EB/OL]. (2017-12-06) [2022-09-19]. . |
16 | SONG Y, WANG J, JIANG T, et al. Attentional encoder network for targeted sentiment classification [EB/OL]. (2019-04-01) [2022-06-19]. . |
17 | PONTIKI M, GALANIS D, PAVLOPOULOS J, et al. SemEval-2014 task 4: aspect based sentiment analysis[C]// Proceedings of the 8th International Workshop on Semantic Evaluation. Stroudsberg: ACL, 2014: 27-35. |
18 | JIANG Q, CHEN L, XU R, et al. A challenge dataset and effective models for aspect-based sentiment analysis[C]// Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing. Stroudsberg: ACL, 2019: 6280-6285. |
19 | DONG L, WEI F, TAN C, et al. Adaptive recursive neural network for target-dependent Twitter sentiment classification[C]// Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics (Volume 2: Short papers). Stroudsberg: ACL, 2014: 49-54. |
20 | DOZAT T, MANNING C D. Deep biaffine attention for neural dependency parsing [EB/OL]. (2017-03-10) [2022-06-19]. . |
21 | JOZEFOWICZ R, ZAREMBA W, SUTSKEVER I. An empirical exploration of recurrent network architectures[C]// Proceedings of the 32nd International Conference on Machine Learning. New York: JMLR.org, 2015: 2342-2350. |
22 | KIRITCHENKO S, ZHU X, CHERRY C, et al. NRD-Canada-2014: detecting aspects and sentiment in customer reviews[C]// Proceedings of the 8th International Workshop on Semantic Evaluation. Stroudsberg: ACL, 2014: 437-442. |
23 | WANG Y, HUANG M, ZHU X, et al. Attention-based LSTM for aspect-level sentiment classification[C]// Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing. Stroudsberg: ACL, 2016: 606-615. |
24 | LI X, BING L, LAM W, et al. Transformation networks for target-oriented sentiment classification[C]// Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). Stroudsberg: ACL, 2018: 946-956. |
25 | NGUYEN T H, SHIRAI K. PhraseRNN: phrase recursive neural network for aspect-based sentiment analysis [C]// Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing. Stroudsberg: ACL, 2015: 2509-2514. |
26 | SUN K, ZHANG R, MENSAH S, et al. Aspect-level sentiment analysis via convolution over dependency tree [C]// Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing. Stroudsberg: ACL, 2019: 5679-5688. |
27 | LI X, LU R, LIU P, et al. Graph convolutional networks with hierarchical multi-head attention for aspect-level sentiment classification [J]. The Journal of Supercomputing, 2022, 78(13):14846-14865. |
28 | LIANG S, WEI W, MAO X-L, et al. BiSyn-GAT+: bi-syntax aware graph attention network for aspect-based sentiment analysis [EB/OL]. (2019-04-01) [2022-06-19]. . |
29 | ZHOU X, ZHANG T, CHENG C, et al. Dynamic multichannel fusion mechanism based on a graph attention network and BERT for aspect-based sentiment classification[J]. Applied Intelligence, 2023, 53(6): 6800-6813. |
[1] | Xianfeng YANG, Yilei TANG, Ziqiang LI. Aspect-level sentiment analysis model based on alternating‑attention mechanism and graph convolutional network [J]. Journal of Computer Applications, 2024, 44(4): 1058-1064. |
[2] | Lei GUO, Zhen JIA, Tianrui LI. Relational and interactive graph attention network for aspect-level sentiment analysis [J]. Journal of Computer Applications, 2024, 44(3): 696-701. |
[3] | Dapeng XU, Xinmin HOU. Feature selection method for graph neural network based on network architecture design [J]. Journal of Computer Applications, 2024, 44(3): 663-670. |
[4] | Linqin WANG, Te ZHANG, Zhihong XU, Yongfeng DONG, Guowei YANG. Fusing entity semantic and structural information for knowledge graph reasoning [J]. Journal of Computer Applications, 2024, 44(11): 3371-3378. |
[5] | Wenjuan JIANG, Yi GUO, Jiaojiao FU. Reasoning question answering model of complex temporal knowledge graph with graph attention [J]. Journal of Computer Applications, 2024, 44(10): 3047-3057. |
[6] | Hongjun HENG, Dingcheng YANG. Knowledge enhanced aspect word interactive graph neural network [J]. Journal of Computer Applications, 2023, 43(8): 2412-2419. |
[7] | Zhixiong ZHENG, Jianhua LIU, Shuihua SUN, Ge XU, Honghui LIN. Aspect-based sentiment analysis model fused with multi-window local information [J]. Journal of Computer Applications, 2023, 43(6): 1796-1802. |
[8] | Jinyun WANG, Yang XIANG. Text semantic de-duplication algorithm based on keyword graph representation [J]. Journal of Computer Applications, 2023, 43(10): 3070-3076. |
[9] | Chun GAO, Mengling WANG. Highway traffic flow prediction based on feature fusion graph attention network [J]. Journal of Computer Applications, 2023, 43(10): 3114-3120. |
[10] | Shigang YANG, Yongguo LIU. Short text classification method by fusing corpus features and graph attention network [J]. Journal of Computer Applications, 2022, 42(5): 1324-1329. |
[11] | Shoulong JIAO, Youxiang DUAN, Qifeng SUN, Zihao ZHUANG, Chenhao SUN. Knowledge representation learning method incorporating entity description information and neighbor node features [J]. Journal of Computer Applications, 2022, 42(4): 1050-1056. |
[12] | Jiana MENG, Pin LYU, Yuhai YU, Shichang SUN, Hongfei LIN. Aspect-level cross-domain sentiment analysis based on capsule network [J]. Journal of Computer Applications, 2022, 42(12): 3700-3707. |
[13] | ZHANG Yang, JIANG Minghu. Authorship identification of text based on attention mechanism [J]. Journal of Computer Applications, 2021, 41(7): 1897-1901. |
[14] | Haitao XUE, Li WANG, Yanjie YANG, Biao LIAN. Rumor detection model based on user propagation network and message content [J]. Journal of Computer Applications, 2021, 41(12): 3540-3545. |
[15] | WANG Jiaxin, FENG Yi, YOU Rui. Network security measurment based on dependency relationship graph and common vulnerability scoring system [J]. Journal of Computer Applications, 2019, 39(6): 1719-1727. |
Viewed | ||||||
Full text |
|
|||||
Abstract |
|
|||||