Please wait a minute...
Advanced Search
数据分析与知识发现  2024, Vol. 8 Issue (1): 55-68     https://doi.org/10.11925/infotech.2096-3467.2022.1259
  研究论文 本期目录 | 过刊浏览 | 高级检索 |
融合抽象语义表示和依存语法的方面级情感分析*
李雪莲1,王碧2(),李立鑫3,韩迪轩4
1南京邮电大学外国语学院 南京 210023
2江西理工大学信息工程学院 赣州 341000
3东南大学计算机科学与工程学院 南京 210096
4佐治亚理工学院电子计算机工程学院 亚特兰大 30318
Sentiment Analysis with Abstract Meaning Representation and Dependency Grammar
Li Xuelian1,Wang Bi2(),Li Lixin3,Han Dixuan4
1School of Foreign Studies, Nanjing University of Posts and Telecommunications, Nanjing 210023, China
2School of Information Engineering, Jiangxi University of Science and Technology, Ganzhou 341000, China
3School of Computer Science and Engineering, Southeast University, Nanjing 210096, China
4School of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta 30318, United States
全文: PDF (959 KB)   HTML ( 16 ) 
输出: BibTeX | EndNote (RIS)      
摘要 

【目的】 充分融合自然语言句子的深层语义表示和表层语法结构，实现语义和语法的互补。【方法】 提出基于规则串联的抽象语义表示和依存语法集成策略，并进行方面级情感分析。该策略利用回答集编程语言（ASP）将抽象语义表示、依存语法、词性分别表示为ASP事实，并基于抽象语义规则，通过规则体扩展的方式集成依存语法和词性，即将一个句子中的多种语言特征依次在规则体中进行串联来使用。该策略实现了两种方法：集成语义、语法和词性信息的AMR-DEP-POS-C方法；忽略词性信息的AMR-DEP-C方法。【结果】 在8个公开评论数据集上的实验表明，AMR-DEP-POS-C方法能够实现语义和语法间的互补，比语义规则方法、语法规则方法以及基于深度学习的方法具有更好性能。【局限】 该方法依赖于现有抽象语义表示和依存语法分析工具的准确性。【结论】 AMR-DEP-POS-C方法具有可解释性，不需要大规模数据集，能够有效融合不同的语言特征，可为方面级情感分析任务带来新的研究视角和工具。
服务
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章
李雪莲
王碧
李立鑫
韩迪轩
关键词 ： 抽象语义表示,  依存语法,  规则,  方面级情感分析    
Abstract：

[Objective] This paper aims to combine the deep semantic representation and surface syntactic structure of natural language sentences. [Methods] We proposed an integration strategy based on semantic and syntactic rule concatenation and utilized it for the aspect-based sentiment analysis. This strategy used the answer set programming language (ASP) to represent abstract meaning representation (AMR), dependency grammar (DEP), and part of speech (POS) as ASP facts. It also integrated the DEP and POS through rule body extension based on AMR rules. Therefore, a sentence’s two or more language features were concatenated into the rule body. Based on this strategy, we developed the AMR-DEP-POS-C and AMR-DEP-C models. [Results] We examined the new methods on eight publicly available review datasets. The AMR-DEP-POS-C achieved a complementary relationship between semantics and syntax and performed better than the baseline methods based on semantic, syntactic, and deep learning. [Limitations] Our new models rely on the accuracy of the existing AMR and DEP parsers. [Conclusions] AMR-DEP-POS-C can effectively integrate different language features and bring new research perspectives and tools for aspect-based sentiment analysis.
Key words： Abstract Meaning Representation    Dependency Grammar    Rule    Aspect-Based Sentiment Analysis
收稿日期: 2022-11-27      出版日期: 2023-04-28
ZTFLH:  TP391  
  H03  
基金资助:*江苏省双创博士(JSSCBS20220624);南京邮电大学人才引进项目(XK0094522034);江西省自然科学基金项目(20232BAB212022)
通讯作者: 王碧，ORCID：0000-0002-4365-0148，E-mail：wangbi@jxust.edu.cn。   
引用本文:   
李雪莲, 王碧, 李立鑫, 韩迪轩. 融合抽象语义表示和依存语法的方面级情感分析*[J]. 数据分析与知识发现, 2024, 8(1): 55-68.
Li Xuelian, Wang Bi, Li Lixin, Han Dixuan. Sentiment Analysis with Abstract Meaning Representation and Dependency Grammar. Data Analysis and Knowledge Discovery, 2024, 8(1): 55-68.
链接本文:  
https://manu44.magtech.com.cn/Jwk_infotech_wk3/CN/10.11925/infotech.2096-3467.2022.1259      或      https://manu44.magtech.com.cn/Jwk_infotech_wk3/CN/Y2024/V8/I1/55
Fig.1  基于规则串联的种子规则构建示意图
类型 ASP事实
词性事实Factpos pos （the， dt）.
pos （camera， nn）.
pos （has， vbz）.
pos （a， dt）.
pos （good， jj）.
pos （battery，nn）.
语法事实Factdep dep（camera， det， the）.
dep（has，nsubj， camera）.
dep（has， dobj， battery）.
dep（battery，det， a）.
dep（battery， amod，good）.
语义事实Factamr amr（battery， mod，good）.
amr（battery，domain， camera）.
观点词事实Factopn opinionword（good）.
Table 1  例句1的ASP事实
编号 名称 句子数量 标注的评价对象数量
D1 Digital Camera 597 237
D2 Digital Camera 346 174
D3 Cell Phone 546 302
D4 MP3 Player 1 716 674
D5 DVD Player 740 296
D6 Restaurant14（Train） 3 044 3 699
Restaurant14（Test） 800 1 134
D7 Laptop14（Train） 3 048 2 373
Laptop14（Test） 800 654
D8 Restaurant15（Train） 1 315 1 279
Restaurant15（Test） 685 597
Table 2  评价对象抽取数据集的统计信息
实验环境 类别 参数
硬件环境 CPU Intel（R） Core（TM） i53210M CPU @2.50GHz
内存 14GB
软件环境 系统 Ubuntu 16.04.4 LTS
编程语言 Python3.6.5 |Anconda
Python 编辑器 VIM Vi IMproved 7.4 & Jupyter Notebook
回答集求解器版本 clingo-4.5.4
Table 3  实验环境信息
方法 评价指标 D1 D2 D3 D4 D5 平均 D6 D7 D8 平均
DP mul 70.7 73.6 76.5 69.7 63.0 70.7 84.3 70.7 78.2 77.8
dis 60.0 59.6 58.1 54.0 52.8 56.9 73.2 67.7 69.4 70.1
DP+ mul 66.2 65.7 65.6 62.2 58.8 63.7 75.2 63.1 69.8 69.4
dis 46.6 46.3 45.9 46.1 45.6 46.1 62.4 59.2 59.1 60.3
RSDP mul 86.2 88.6 84.2 81.0 84.1 84.8 86.9 83.0 86.6 85.5
dis 82.7 82.8 76.1 72.0 74.5 77.6 78.5 79.7 81.9 80.1
RSDP+ mul 85.2 87.0 83.3 80.8 85.5 84.4 88.7 83.1 87.5 86.4
dis 80.9 83.1 76.3 68.5 78.3 77.4 82.0 80.1 83.2 81.8
DEP mul 75.5 86.2 75.0 72.1 71.8 76.1 88.4 65.3 74.2 76.0
dis 67.0 78.3 64.0 59.1 60.6 65.8 79.3 61.6 64.4 68.4
AMR mul 69.2 79.3 76.0 68.3 71.7 72.9 72.2 67.5 71.6 70.4
dis 60.6 68.5 65.7 56.4 68.9 64.0 59.4 65.8 56.1 60.4
CRF mul 79.3 84.0 80.0 87.0 74.4 80.9 - - - -
dis 63.8 73.5 70.6 80.2 65.6 70.7 78.3 62.1 - -
AMR-POS mul 76.0 84.2 76.6 75.3 79.1 78.2 83.8 64.4 85.2 77.8
dis 70.3 75.2 67.8 67.3 71.5 70.4 72.7 62.3 79.3 71.4
AMR-DEP-C mul 80.5 88.5 87.8 71.9 69.4 79.6 81.9 81.7 76.9 80.2
dis 75.3 81.0 80.5 57.9 64.1 71.7 68.8 75.9 63.9 69.5
AMR-DEP-POS-C mul 83.8 89.0 85.8 81.7 87.2 85.5 90.2 87.6 91.5 89.8
dis 77.3 83.2 79.2 72.0 83.9 79.1 82.7 82.9 86.5 84.0
Table 4  本文方法与基线方法在D1-D8数据集上的准确率（%）
方法 评价指标 D1 D2 D3 D4 D5 平均 D6 D7 D8 平均
DP mul 91.0 89.5 90.2 88.7 89.6 89.8 86.0 75.4 89.9 83.8
dis 83.9 78.8 81.4 74.7 76.3 79.0 74.3 69.4 83.1 75.6
DP+ mul 96.3 95.9 95.1 95.6 94.3 95.4 92.6 89.5 95.2 92.4
dis 91.4 89.4 87.6 88.0 87.1 88.7 86.4 84.2 92.0 87.5
RSDP mul 88.1 87.8 82.9 81.5 78.9 83.8 82.7 69.2 83.2 78.4
dis 74.2 75.8 67.0 66.0 58.1 68.2 68.9 61.7 73.1 67.9
RSDP+ mul 90.5 93.6 89.5 90.4 90.0 90.8 84.9 69.8 84.9 79.9
dis 75.3 83.3 73.2 78.0 67.7 75.5 71.6 62.5 75.9 70.0
DEP mul 87.7 93.7 82.8 88.7 84.0 87.4 78.7 79.4 78.4 78.8
dis 78.5 83.6 71.4 74.7 61.7 74.0 60.6 72.1 65.8 66.2
AMR mul 63.5 80.5 76.6 87.1 75.8 76.7 81.5 58.0 85.4 75.0
dis 53.8 64.2 64.3 71.3 46.8 60.1 65.6 51.0 71.8 62.8
CRF mul 67.9 74.4 51.1 70.3 69.9 66.7 - - - -
dis 52.7 54.6 45.4 50.0 50.5 50.6 71.6 57.8 - -
AMR-POS mul 95.5 94.8 93.1 90.8 93.2 93.5 85.0 73.6 75.8 78.1
dis 90.3 88.1 82.7 81.3 86.2 85.8 71.8 68.1 59.8 66.6
AMR-DEP-C mul 90.2 92.5 90.3 90.8 85.4 89.9 87.9 73.3 87.8 83.0
dis 78.5 80.6 75.5 77.3 63.8 75.2 76.4 65.8 77.5 73.2
AMR-DEP-POS-C mul 93.4 96.6 87.2 89.2 90.4 91.4 87.7 72.1 86.7 82.2
dis 86.0 91.0 75.5 78.0 79.8 82.1 76.4 64.8 75.9 72.4
Table 5  本文方法与基线方法在D1-D8数据集上的召回率（%）
方法 评价指标 D1 D2 D3 D4 D5 平均 D6 D7 D8 平均
DP mul 80.0 80.8 82.8 78.1 74.0 79.1 85.2 73.0 83.6 80.6
dis 70.0 67.9 67.9 62.6 62.4 66.1 73.7 68.5 75.6 72.6
DP+ mul 78.5 78.0 77.6 75.4 72.4 76.4 83.0 74.0 80.5 79.2
dis 61.7 61.0 60.3 60.5 59.8 60.7 72.5 69.5 72.0 71.3
RSDP mul 87.1 88.2 83.5 81.2 81.4 84.3 84.8 75.5 84.9 81.7
dis 78.2 79.1 71.3 68.8 65.3 72.5 73.4 69.6 77.3 73.4
RSDP+ mul 87.8 90.2 86.3 85.3 87.7 87.4 86.7 75.9 86.2 82.9
dis 78.0 83.2 74.7 73.0 72.7 76.3 76.4 70.2 79.4 75.3
DEP mul 81.2 89.8 78.7 79.6 77.4 81.3 83.3 71.7 76.2 77.1
dis 72.3 80.8 67.5 66.0 61.2 69.6 68.7 66.5 65.1 66.8
AMR mul 66.2 79.9 76.3 76.6 73.7 74.5 76.5 62.4 77.9 72.3
dis 57.0 66.3 65.0 63.0 55.7 61.4 62.4 57.5 63.0 61.0
CRF mul 73.1 78.9 62.3 77.8 72.1 72.9 - - - -
dis 57.7 62.6 55.2 61.6 57.1 58.9 74.8 59.9 - -
AMR-POS mul 84.7 89.2 84.0 82.3 85.6 85.2 84.4 68.7 80.2 77.8
dis 79.0 81.1 74.5 73.7 78.2 77.3 72.3 65.1 68.2 68.5
AMR-DEP-C mul 85.1 90.5 89.0 80.2 76.6 84.3 84.8 77.2 82.0 81.3
dis 76.9 80.8 77.9 66.2 64.0 73.1 72.4 70.4 70.0 70.9
AMR-DEP-POS-C mul 88.4 92.6 86.5 85.3 88.8 88.3 88.9 79.1 89.0 85.7
dis 81.4 86.9 77.3 74.9 81.8 80.5 79.4 72.8 80.8 77.7
Table 6  本文方法与基线方法在D1-D8数据集上的F1值（%）
Fig.2  AMR-DEP-POS-C与R-DNN、CRF的性能比较
方法 最终规则
数量		 准确率
（%）		 召回率
（%）		 F1值
（%）
RSDP+ 130 82.0 71.6 76.4
AMR-POS 227 72.7 71.8 72.3
AMR-DEP-C 360 68.8 76.4 72.4
AMR-DEP-POS-C 990 82.7 76.4 79.4
Table 7  4种方法的规则数量和性能比较
方法 实例化
规则数量		 候选规则
数量		 最终规则
数量		 运行时间
（min）
RSDP+ 313 215 130 44
AMR-POS 613 329 227 61
AMR-DEP-C 1 528 559 360 113
AMR-DEP-POS-C 3 725 1 976 990 259
Table 8  4种方法运行时间和规则数量
编号 RSDP+ AMR-POS AMR-DEP-POS-C 评价对象数量
1 - - - 95
2 - - + 13
3 - + - 31
4 - + + 18
5 + - - 15
6 + - + 33
7 + + - 9
8 + + + 309
Table 9  三种方法的评价对象抽取结果对比
编号 错误类型 具体类别 错误评价
对象占比
1 第一类
错误		 词性标注错误 25%
2 依存语法解析错误
3 AMR解析错误
4 第二类
错误		 评价对象与观点词无一阶语法关系 63%
5 评价对象与观点词无二阶语法关系
6 评价对象之间无一阶语法关系
7 评价对象之间无二阶语法关系
8 评价对象与观点词无一阶语义关系
9 评价对象与观点词无二阶语义关系
10 评价对象之间无一阶语义关系
11 评价对象之间无二阶语义关系
12 第三类错误 第三类错误约束条件多 12%
Table 10  AMR-DEP-POS-C方法中评价对象抽取错误类型分布
[1] Pang B, Lee L. Opinion Mining and Sentiment Analysis[J]. Foundations and Trends in Information Retrieval, 2008, 2(1-2): 1-135.
doi: 10.1561/1500000011
[2] 谭荧, 张进, 夏立新. 社交媒体情境下的情感分析研究综述[J]. 数据分析与知识发现, 2020, 4(1): 1-11.
[2] (Tan Ying, Zhang Jin, Xia Lixin. A Survey of Sentiment Analysis on Social Media[J]. Data Analysis and Knowledge Discovery, 2020, 4(1): 1-11.)
[3] 钟佳娃, 刘巍, 王思丽, 等. 文本情感分析方法及应用综述[J]. 数据分析与知识发现, 2021, 5(6): 1-13.
[3] (Zhong Jiawa, Liu Wei, Wang Sili, et al. Review of Methods and Applications of Text Sentiment Analysis[J]. Data Analysis and Knowledge Discovery, 2021, 5(6): 1-13.)
[4] 商容轩, 张斌, 米加宁. 基于BRNN的政务APP评论端到端方面级情感分析方法[J]. 数据分析与知识发现, 2022, 6(2/3): 364-375.
[4] (Shang Rongxuan, Zhang Bin, Mi Jianing. End-to-End Aspect-Level Sentiment Analysis for E-Government Applications Based on BRNN[J]. Data Analysis and Knowledge Discovery, 2022, 6(2/3): 364-375.)
[5] Rastegar-Mojarad M, Elayavilli R K, Wang Y S, et al. Semantic Information Retrieval: Exploring Dependency and Word Embedding Features in Biomedical Information Retrieval[C]// Proceedings of the 2017 BioCreative VI Challenge Evaluation Workshop. 2017: 74-77.
[6] Wang Y S, Liu S J, Rastegar-Mojarad M, et al. Dependency and AMR Embeddings for Drug-Drug Interaction Extraction from Biomedical Literature[C]// Proceedings of the 8th ACM International Conference on Bioinformatics, Computational Biology, and Health Informatics. ACM, 2017: 36-43.
[7] Marcheggiani D, Bastings J, Titov I. Exploiting Semantics in Neural Machine Translation with Graph Convolutional Networks[C]// Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics:Human Language Technologies. 2018: 486-492.
[8] Gebser M, Kaminski R, Kaufmann B, et al. Answer Set Solving in Practice[M]. Cham: Springer, 2012.
[9] Son T C, Baral C, Tran N, et al. Domain-Dependent Knowledge in Answer Set Planning[J]. ACM Transactions on Computational Logic, 2002, 7(4): 613-657.
doi: 10.1145/1183278.1183279
[10] Baral C. Knowledge Representation, Reasoning, and Declarative Problem Solving[M]. Cambridge: Cambridge University Press, 2003.
[11] van der Hallen M, Paramonov S, Janssens G, et al. Knowledge Representation Analysis of Graph Mining[J]. Annals of Mathematics and Artificial Intelligence, 2019, 86(1): 21-60.
doi: 10.1007/s10472-019-09624-y
[12] Falkner A, Friedrich G, Schekotihin K, et al. Industrial Applications of Answer Set Programming[J]. KI - Künstliche Intelligenz, 2018, 32(2): 165-176.
doi: 10.1007/s13218-018-0548-6
[13] Wang Y, Lee J, Kim D S. A Logic Based Approach to Answering Questions About Alternatives in DIY Domains[C]// Proceedings of the 31st AAAI Conference on Artificial Intelligence. ACM, 2017: 4753-4758.
[14] Piovesan L, Terenziani P, Theseider Dupré D. Conformance Analysis for Comorbid Patients in Answer Set Programming[J]. Journal of Biomedical Informatics, 2020, 103: Article No.103377.
[15] Gebser M, Kaufmann B, Schaub T. Conflict-Driven Answer Set Solving: From Theory to Practice[J]. Artificial Intelligence, 2012, 187-188: 52-89.
[16] Faber W, Pfeifer G, Leone N, et al. Design and Implementation of Aggregate Functions in the DLV System[J]. Theory and Practice of Logic Programming, 2008, 8(5-6): 545-580.
doi: 10.1017/S1471068408003323
[17] Simons P, Niemelä I, Soininen T. Extending and Implementing the Stable Model Semantics[J]. Artificial Intelligence, 2002, 138(1-2): 181-234.
doi: 10.1016/S0004-3702(02)00187-X
[18] Liu B. Sentiment Analysis and Opinion Mining[M]. Cham: Springer, 2012.
[19] Qiu G, Liu B, Bu J J, et al. Opinion Word Expansion and Target Extraction Through Double Propagation[J]. Computational Linguistics, 2011, 37(1): 9-27.
doi: 10.1162/coli_a_00034
[20] Liu Q, Gao Z Q, Liu B, et al. A Logic Programming Approach to Aspect Extraction in Opinion Mining[C]// Proceedings of the 2013 IEEE/WIC/ACM International Joint Conferences on Web Intelligence and Intelligent Agent Technologies. IEEE, 2013: 276-283.
[21] Liu Q, Gao Z Q, Liu B, et al. Automated Rule Selection for Aspect Extraction in Opinion Mining[C]// Proceedings of the 24th International Conference on Artificial Intelligence. ACM, 2015: 1291-1297.
[22] Liu Q, Liu B, Zhang Y L, et al. Improving Opinion Aspect Extraction Using Semantic Similarity and Aspect Associations[C]// Proceedings of the 30th AAAI Conference on Artificial Intelligence. ACM, 2016: 2986-2992.
[23] 王铁琨. 基于语言资源理念的语言规划——以“语言资源监测研究” 和“中国语言资源有声数据库建设” 为例[J]. 陕西师范大学学报(哲学社会科学版), 2010, 39(6): 58-66.
[23] Wang Tiekun. Language Planning Based on the Conception of Language Resources—An Exemplification of “Monitoring Study of Language Resources” and “Construction of the Corpus of Vocal Data of Chinese Language Resources”[J]. Journal of Shaanxi Normal University (Philosophy and Social Sciences Edition), 2010, 39(6): 58-66.)
[24] 王世凯. 语言资源与语言研究[M]. 上海: 学林出版社, 2009.
[24] (Wang Shikai. Language Resources and Language Studies[M]. Shanghai: Academia Press, 2009.)
[25] Liu Q, Gao Z Q, Liu B, et al. Automated Rule Selection for Opinion Target Extraction[J]. Knowledge-Based Systems, 2016, 104: 74-88.
doi: 10.1016/j.knosys.2016.04.010
[26] Zhang S, Jia W J, Xia Y J, et al. Opinion Analysis of Product Reviews[C]// Proceedings of the 6th International Conference on Fuzzy Systems and Knowledge Discovery. IEEE, 2009: 591-595.
[27] 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. ACM, 2009: 465-472.
[28] Yu J X, Zha Z J, Wang M, et al. Aspect Ranking: Identifying Important Product Aspects from Online Consumer Reviews[C]//Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics:Human Language Technologies. ACM, 2011: 1496-1505.
[29] Ma D H, Li S J, Wu F Z, et al. Exploring Sequence-to-Sequence Learning in Aspect Term Extraction[C]// Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics. 2019: 3538-3547.
[30] Wu C H, Wu F Z, Wu S X, et al. A Hybrid Unsupervised Method for Aspect Term and Opinion Target Extraction[J]. Knowledge-Based Systems, 2018, 148: 66-73.
doi: 10.1016/j.knosys.2018.01.019
[31] Hu M Q, Liu B. Mining and Summarizing Customer Reviews[C]// Proceedings of the 10th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. ACM, 2004: 168-177.
[32] Zhuang L, Jing F, Zhu X Y. Movie Review Mining and Summarization[C]// Proceedings of the 15th ACM International Conference on Information and Knowledge Management. ACM, 2006: 43-50.
[33] Wang B, Wang H F. Bootstrapping Both Product Features and Opinion Words from Chinese Customer Reviews with Cross-Inducing[C]// Proceedings of the 3rd International Joint Conference on Natural Language Processing. 2008: 289-295.
[34] Wu Y B, Zhang Q, Huang X J, et al. Phrase Dependency Parsing for Opinion Mining[C]// Proceedings of the Conference on Empirical Methods in Natural Language Processing. ACM, 2009: 1533-1541.
[35] Zhang L, Liu B, Lim S H, et al. Extracting and Ranking Product Features in Opinion Documents[C]//Proceedings of the 23rd International Conference on Computational Linguistics:Posters. ACM, 2010: 1462-1470.
[36] Zhao Y Y, Qin B, Hu S, et al. Generalizing Syntactic Structures for Product Attribute Candidate Extraction[C]// Proceedings of the Annual Conference of the North American Chapter of the Association for Computational Linguistics. ACM, 2010: 377-380.
[37] Wang W Y, Pan S J, Dahlmeier D, et al. Recursive Neural Conditional Random Fields for Aspect-Based Sentiment Analysis[C]// Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing. 2016: 616-626.
[38] Yin Y C, Wei F R, Dong L, et al. Unsupervised Word and Dependency Path Embeddings for Aspect Term Extraction[C]// Proceedings of the 25th International Joint Conference on Artificial Intelligence. ACM, 2016: 2979-2985.
[39] Ye H, Yan Z C, Luo Z C, et al. Dependency-Tree Based Convolutional Neural Networks for Aspect Term Extraction[C]// Proceedings of the 2017 Pacific-Asia Conference on Knowledge Discovery and Data Mining. Cham: Springer, 2017: 350-362.
[40] Wang W Y, Pan S J, Dahlmeier D. Memory Networks for Fine-Grained Opinion Mining[J]. Artificial Intelligence, 2018, 265: 1-17.
doi: 10.1016/j.artint.2018.09.002
[41] 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.
[42] Li X, Bing L D, Li P J, et al. A Unified Model for Opinion Target Extraction and Target Sentiment Prediction[C]// Proceedings of the 33rd Conference on Artificial Intelligence and 31st Innovative Applications of Artificial Intelligence Conference and 9th AAAI Symposium on Educational Advances in Artificial Intelligence. 2019: 6714-6721.
[43] Do H H, Prasad P, Maag A, et al. Deep Learning for Aspect-Based Sentiment Analysis: A Comparative Review[J]. Expert Systems with Applications, 2019, 118: 272-299.
doi: 10.1016/j.eswa.2018.10.003
[44] Liao C, Feng C, Yang S, et al. A Hybrid Method of Domain Lexicon Construction for Opinion Targets Extraction Using Syntax and Semantics[J]. Journal of Computer Science and Technology, 2016, 31(3): 595-603.
doi: 10.1007/s11390-016-1649-z
[45] Kabir M, Everardo F O, Shukla A K, et al. ApproxASP—A Scalable Approximate Answer Set Counter[C]// Proceedings of the 36th AAAI Conference on Artificial Intelligence. 2022: 5755-5764.
[46] Obermeier P. Scalable Robotic Intra-Logistics with Answer Set Programming[C]// Proceedings of the 34th International Conference on Logic Programming. 2018(24): 5-9.
[47] Jakob N, Gurevych I. Extracting Opinion Targets in a Single- and Cross-Domain Setting with Conditional Random Fields[C]// Proceedings of the 2010 Conference on Empirical Methods in Natural Language Processing. ACM, 2010: 1035-1045.
[1] 赵又霖, 徐竟楠, 陆颖隽. 面向实时财经信息的领域情感歧义搭配词典构建研究*[J]. 数据分析与知识发现, 2023, 7(7): 100-110.
[2] 商容轩, 张斌, 米加宁. 基于BRNN的政务APP评论端到端方面级情感分析方法*[J]. 数据分析与知识发现, 2022, 6(2/3): 364-375.
[3] 盛羽, 胡慧荣, 王聪聪, 杨晟艺. 医学影像诊断报告的结构化研究*[J]. 数据分析与知识发现, 2022, 6(10): 46-56.
[4] 李铁军,颜端武,杨雄飞. 基于情感加权关联规则的微博推荐研究*[J]. 数据分析与知识发现, 2020, 4(4): 27-33.
[5] 魏伟,郭崇慧,邢小宇. 基于语义关联规则的试题知识点标注及试题推荐*[J]. 数据分析与知识发现, 2020, 4(2/3): 182-191.
[6] 黄名选,卢守东,徐辉. 基于加权关联模式挖掘与规则后件扩展的跨语言信息检索 *[J]. 数据分析与知识发现, 2019, 3(9): 77-87.
[7] 强韶华,罗云鹿,李玉鹏,吴鹏. 基于RBR和CBR的金融事件本体推理研究 *[J]. 数据分析与知识发现, 2019, 3(8): 94-104.
[8] 张勇,李树青,程永上. 基于频次有效长度的加权关联规则挖掘算法研究 *[J]. 数据分析与知识发现, 2019, 3(7): 85-93.
[9] 卢强,朱振方,徐富永,国强强. 融合语法规则的Bi-LSTM中文情感分类方法研究 *[J]. 数据分析与知识发现, 2019, 3(11): 99-107.
[10] 何跃, 丰月, 赵书朋, 马玉凤. 基于知乎问答社区的内容推荐研究——以物流话题为例[J]. 数据分析与知识发现, 2018, 2(9): 42-49.
[11] 何跃, 王爱欣, 丰月, 王莉. 基于关联规则的门诊药房布局优化[J]. 数据分析与知识发现, 2018, 2(1): 99-108.
[12] 魏星, 胡德华, 易敏寒, 朱启贞, 朱文婕. 基于数据立方体挖掘疾病-基因-药物新关联*[J]. 数据分析与知识发现, 2017, 1(10): 94-104.
[13] 黄名选. 基于矩阵加权关联模式的印尼中跨语言信息检索模型*[J]. 数据分析与知识发现, 2017, 1(1): 26-36.
[14] 马天翼,张朋柱,冯浩垠. 网络外包任务的知识需求建模研究*[J]. 现代图书情报技术, 2016, 32(3): 74-81.
[15] 阮光册, 夏磊. 基于关联规则的文本主题深度挖掘应用研究*[J]. 数据分析与知识发现, 2016, 32(12): 50-56.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
版权所有 © 2015 《数据分析与知识发现》编辑部
地址：北京市海淀区中关村北四环西路33号 邮编：100190
电话/传真：(010)82626611-6626，82624938
E-mail:jishu@mail.las.ac.cn