maisМоделирование и анализ информационных системModeling and Analysis of Information Systems1818-10152313-5417Yaroslavl State University10.18255/1818-1015-2024-4-362-383mais-1895Research ArticleArtificial IntelligenceМетоды определения тональности по отношению к аспектам социально-экономического развития в предложениях на русском языкеMethods of sentiment detection towards aspect of economic and social development in Russian sentenceshttps://orcid.org/0000-0003-0116-4739ПолетаевАнатолий ЮрьевичPoletaevAnatoliy Y.anatoliy-poletaev@mail.ruhttps://orcid.org/0000-0003-3984-8423ПарамоновИлья ВячеславовичParamonovIlya V.ilya.paramonov@fruct.orghttps://orcid.org/0000-0001-6600-2971БойчукЕлена ИгоревнаBoychukElena I.elena-boychouk@rambler.ruЯрославский государственный университет им. П.Г. ДемидоваРоссияP.G. Demidov Yaroslavl State UniversityRussian Federation202413122024314362383Copyright © Полетаев А.Ю., Парамонов И.В., Бойчук Е.И., 20242024Полетаев А.Ю., Парамонов И.В., Бойчук Е.И.Poletaev A.Y., Paramonov I.V., Boychuk E.I.Данная работа распространяется под лицензией Creative Commons Attribution 4.0.This work is licensed under a Creative Commons Attribution 4.0 License.https://www.mais-journal.ru/jour/article/view/1895

Статья посвящена задаче определения тональности по отношению к аспектам социально-экономического развития в предложениях на русском языке. Аспект, отношение к которому определяется, может как упоминаться явно, так и подразумеваться. Авторами были исследованы возможности применения нейросетевых классификаторов, а также предложен алгоритм определения тональности по отношению к аспекту, основанный на семантических правилах, реализованных с использованием деревьев синтаксических единиц. Тональность по отношению к аспекту определяется в два этапа. На первом этапе в предложении отыскиваются аспектные термины — явно упоминаемые события или явления, связанные с аспектом. На втором этапе тональность по отношению к аспекту определяется как тональность по отношению к аспектному термину, который теснее всего связан с аспектом. В работе предлагается несколько методов поиска аспектных терминов. Качество оценивалось на корпусе из 468 предложений, извлечённых из материалов предвыборной агитации. Лучший результат для нейросетевых классификаторов был получен с использованием нейронной сети BERT-SPC, предобученной на задаче определения тональности по отношению к явно упоминаемому аспекту, макро-F-мера составила 0.74. Лучший результат для алгоритма, основанного на семантических правилах, был получен при использовании метода поиска аспектных терминов на основе семантической схожести, макро-F-мера составила 0.63. При объединении BERT-SPC и алгоритма, основанного на правилах, в ансамбль была получена макро-F-мера, равная 0.79, что является лучшим результатом, полученным в рамках работы.

The article is devoted to the task of the sentiment detection towards an aspect of economic and social development in Russian sentences. The aspect, the attitude to which is determined, can be either explicitly mentioned or implied. The authors investigated possibilities of using neural network classifiers and proposed an algorithm for determining the sentiment towards an aspect based on semantic rules implemented with the use of constituency trees. The sentiment towards an aspect is determined in two stages. At the first stage, aspect terms (explicitly mentioned events or phenomena associated with the aspect) are found in the sentence. At the second stage, the sentiment towards an aspect is calculated as the sentiment towards the aspect term that is most closely associated with the aspect. The paper proposes several methods for searching the aspect terms. The performance was assessed on a corpus of 468 sentences extracted from election campaign materials. The best result for neural network classifiers was obtained using the BERT-SPC neural network pretrained on the task of identifying the sentiment towards an explicitly mentioned aspect, the macro F-score was 0.74. The best result for the semantic rule-based algorithm was obtained using the method of aspect term searching based on semantic similarity, the macro-F-score was 0.63. When combining BERT-SPC and the rule-based algorithm into an ensemble, the macro-F-score was 0.79, which is the best result obtained in this work.

анализ тональностиопределение тональноститональность по отношению к аспектунеявно упоминаемые аспектысемантические правилапублицистический стильдерево синтаксических единицsentiment analysissentiment detectionsentiment towards an aspectimplicit aspectsemantic rulespublicismconstituency treeРоссийский научный фонд (проект No 23-21-00495).Russian Science Foundation (Project no. 23-21-00495).ReferencesB. Liu, Sentiment Analysis and Opinion Mining. Springer, 2022.B. Liu, Sentiment Analysis and Opinion Mining. Springer, 2022.W. Zhang, X. Li, Y. Deng, L. Bing, and W. Lam, “A survey on aspect-based sentiment analysis: Tasks, methods, and challenges,” IEEE Transactions on Knowledge and Data Engineering, vol. 35, no. 11, pp. 11019–11038, 2022, doi: 10.1109/TKDE.2022.3230975.W. Zhang, X. Li, Y. Deng, L. Bing, and W. Lam, “A survey on aspect-based sentiment analysis: Tasks, methods, and challenges,” IEEE Transactions on Knowledge and Data Engineering, vol. 35, no. 11, pp. 11019–11038, 2022, doi: 10.1109/TKDE.2022.3230975.O. Alqaryouti and others, “Aspect-based sentiment analysis using smart government review data,” Applied Computing and Informatics, vol. 20, no. 1/2, pp. 142–161, 2024, doi: 10.1016/j.aci.2019.11.003.O. Alqaryouti and others, “Aspect-based sentiment analysis using smart government review data,” Applied Computing and Informatics, vol. 20, no. 1/2, pp. 142–161, 2024, doi: 10.1016/j.aci.2019.11.003.A. Nazir, Y. Rao, L. Wu, and L. Sun, “Issues and challenges of aspect-based sentiment analysis: A comprehensive survey,” IEEE Transactions on Affective Computing, vol. 13, no. 2, pp. 845–863, 2020, doi: 10.1109/TAFFC.2020.2970399.A. Nazir, Y. Rao, L. Wu, and L. Sun, “Issues and challenges of aspect-based sentiment analysis: A comprehensive survey,” IEEE Transactions on Affective Computing, vol. 13, no. 2, pp. 845–863, 2020, doi: 10.1109/TAFFC.2020.2970399.A. Poletaev, I. Paramonov, and E. Kolupaev, “Methods of implicit aspect detection in Russian publicism sentences,” Modeling and Analysis of Information Systems, vol. 31, no. 3, pp. 226–239, 2024, doi: 10.18255/1818-1015-2024-3-226-239.A. Poletaev, I. Paramonov, and E. Kolupaev, “Methods of implicit aspect detection in Russian publicism sentences,” Modeling and Analysis of Information Systems, vol. 31, no. 3, pp. 226–239, 2024, doi: 10.18255/1818-1015-2024-3-226-239.A. D. Kazun, “Construction of social problems in the media and agenda-setting theory: the limits of concepts' compatibility,” Monitoring of Public Opinion: Economic and Social Changes, no. 3 (133), pp. 159–172, 2016, doi: 10.14515/monitoring.2016.3.09.A. D. Kazun, “Construction of social problems in the media and agenda-setting theory: the limits of concepts' compatibility,” Monitoring of Public Opinion: Economic and Social Changes, no. 3 (133), pp. 159–172, 2016, doi: 10.14515/monitoring.2016.3.09.A. I. Guseva, I. A. Kuznetsov, P. V. Bochkarev, and D. S. Smirnov, “DIGITAL SHADOW OF Russian INTERNATIONAL MEGAPROJECTSB OF NPP CONSTRUCTION ABROAD: ASSESSMENT OF THE TONE OF UTTERANCES,” Modern High Technologies, no. 12 (1), pp. 26–34, 2022, doi: 10.17513/snt.39432.A. I. Guseva, I. A. Kuznetsov, P. V. Bochkarev, and D. S. Smirnov, “DIGITAL SHADOW OF Russian INTERNATIONAL MEGAPROJECTSB OF NPP CONSTRUCTION ABROAD: ASSESSMENT OF THE TONE OF UTTERANCES,” Modern High Technologies, no. 12 (1), pp. 26–34, 2022, doi: 10.17513/snt.39432.W. Zhang, X. Li, Y. Deng, L. Bing, and W. Lam, “A Survey on Aspect-Based Sentiment Analysis: Tasks, Methods, and Challenges,” IEEE Transactions on Knowledge and Data Engineering, vol. 35, no. 11, pp. 11019–11038, 2023, doi: 10.1109/TKDE.2022.3230975.W. Zhang, X. Li, Y. Deng, L. Bing, and W. Lam, “A Survey on Aspect-Based Sentiment Analysis: Tasks, Methods, and Challenges,” IEEE Transactions on Knowledge and Data Engineering, vol. 35, no. 11, pp. 11019–11038, 2023, doi: 10.1109/TKDE.2022.3230975.E. G. Brunova, Y. V. Bidulya, and A. A. Gorbunov, “Aspect-based sentiment analysis of political discourse,” Tyumen State University Herald. Humanities Research. Humanitates, vol. 7, no. 3 (27), pp. 6–22, 2021, doi: 10.21684/2411-197X-2021-7-3-6-22.E. G. Brunova, Y. V. Bidulya, and A. A. Gorbunov, “Aspect-based sentiment analysis of political discourse,” Tyumen State University Herald. Humanities Research. Humanitates, vol. 7, no. 3 (27), pp. 6–22, 2021, doi: 10.21684/2411-197X-2021-7-3-6-22.Muljono, B. Harjo, and R. Abdullah, “Aspect-Based Sentiment Analysis for Financial Review with Implicit Aspect and Opinion Using Semantic Similarity and Hybrid Approach,” International Journal of Intelligent Engineering & Systems, vol. 17, no. 5, pp. 646–658, 2024, doi: 10.22266/ijies2024.1031.49.Muljono, B. Harjo, and R. Abdullah, “Aspect-Based Sentiment Analysis for Financial Review with Implicit Aspect and Opinion Using Semantic Similarity and Hybrid Approach,” International Journal of Intelligent Engineering & Systems, vol. 17, no. 5, pp. 646–658, 2024, doi: 10.22266/ijies2024.1031.49.K. Ananthajothi, K. Karthikayani, and R. Prabha, “Explicit and implicit oriented Aspect-Based Sentiment Analysis with optimal feature selection and deep learning for demonetization in India,” Data & Knowledge Engineering, vol. 142, p. 102092, 2022, doi: 10.1016/j.datak.2022.102092.K. Ananthajothi, K. Karthikayani, and R. Prabha, “Explicit and implicit oriented Aspect-Based Sentiment Analysis with optimal feature selection and deep learning for demonetization in India,” Data & Knowledge Engineering, vol. 142, p. 102092, 2022, doi: 10.1016/j.datak.2022.102092.C. Hutto and E. Gilbert, “VADER: A parsimonious rule-based model for sentiment analysis of social media text,” in Proceedings of the International AAAI Conference on Web and Social Media, 2014, vol. 8, no. 1, pp. 216–225.C. Hutto and E. Gilbert, “VADER: A parsimonious rule-based model for sentiment analysis of social media text,” in Proceedings of the International AAAI Conference on Web and Social Media, 2014, vol. 8, no. 1, pp. 216–225.N. Chechneva, “Simple and Efficient Approach to the Aspect Extraction from Customers' Product Reviews,” in Proceedings of the 26th Conference of Open Innovations Association FRUCT, 2020, pp. 67–73, doi: 10.23919/fruct48808.2020.9087546.N. Chechneva, “Simple and Efficient Approach to the Aspect Extraction from Customers' Product Reviews,” in Proceedings of the 26th Conference of Open Innovations Association FRUCT, 2020, pp. 67–73, doi: 10.23919/fruct48808.2020.9087546.A. Poletaev, I. Paramonov, and E. Boychuk, “Automatic Detection of Sentiment Towards Explicit Aspect in Russian Publicism Sentences Using Syntactic Structure,” in Proceedings of the 36th Conference of Open Innovations Association FRUCT, 2024, pp. 593–602.A. Poletaev, I. Paramonov, and E. Boychuk, “Automatic Detection of Sentiment Towards Explicit Aspect in Russian Publicism Sentences Using Syntactic Structure,” in Proceedings of the 36th Conference of Open Innovations Association FRUCT, 2024, pp. 593–602.M. A. Pil'gun, “Rechevye osobennosti politicheskoj kommunikacii,” Proceedings of Kazan University. Humanities Sciences Series, vol. 152, no. 2, pp. 236–246, 2010.M. A. Pil'gun, “Rechevye osobennosti politicheskoj kommunikacii,” Proceedings of Kazan University. Humanities Sciences Series, vol. 152, no. 2, pp. 236–246, 2010.Y. Song and others, “Targeted Sentiment Classification with Attentional Encoder Network,” in Artificial Neural Networks and Machine Learning – ICANN 2019: Text and Time Series, 2019, pp. 93–103, doi: 10.1007/978-3-030-30490-4_9.Y. Song and others, “Targeted Sentiment Classification with Attentional Encoder Network,” in Artificial Neural Networks and Machine Learning – ICANN 2019: Text and Time Series, 2019, pp. 93–103, doi: 10.1007/978-3-030-30490-4_9.J. Ansel and others, “PyTorch 2: Faster Machine Learning Through Dynamic Python Bytecode Transformation and Graph Compilation,” in ASPLOC'24: Proceedings of the 29th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 2, 2024, pp. 929–947, doi: 10.1145/3620665.3640366.J. Ansel and others, “PyTorch 2: Faster Machine Learning Through Dynamic Python Bytecode Transformation and Graph Compilation,” in ASPLOC'24: Proceedings of the 29th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 2, 2024, pp. 929–947, doi: 10.1145/3620665.3640366.A. Naumov and others, “Neural-network method for determining text author’s sentiment to an aspect specified by the named entity,” in CEUR Workshop Proceedings, 2020, vol. 2648, pp. 134–143.A. Naumov and others, “Neural-network method for determining text author’s sentiment to an aspect specified by the named entity,” in CEUR Workshop Proceedings, 2020, vol. 2648, pp. 134–143.Y. Wang, L. Wu, J. Li, X. Liang, and M. Zhang, “Are the BERT family zero-shot learners? A study on their potential and limitations,” Artificial Intelligence, vol. 322, p. 103953, 2023, doi: 10.1016/j.artint.2023.103953.Y. Wang, L. Wu, J. Li, X. Liang, and M. Zhang, “Are the BERT family zero-shot learners? A study on their potential and limitations,” Artificial Intelligence, vol. 322, p. 103953, 2023, doi: 10.1016/j.artint.2023.103953.A. Golubev, N. Rusnachenko, and N. Loukachevitch, “RuSentNE-2023: Evaluating Entity-Oriented Sentiment Analysis on Russian News Texts,” in Computational Linguistics and Intellectual Technologies: Papers from the Annual conference “Dialogue” (2023), 2023, vol. 22, pp. 130–141, doi: 10.28995/2075-7182-2023-22-130-141.A. Golubev, N. Rusnachenko, and N. Loukachevitch, “RuSentNE-2023: Evaluating Entity-Oriented Sentiment Analysis on Russian News Texts,” in Computational Linguistics and Intellectual Technologies: Papers from the Annual conference “Dialogue” (2023), 2023, vol. 22, pp. 130–141, doi: 10.28995/2075-7182-2023-22-130-141.D. Ma and others, “Interactive attention networks for aspect-level sentiment classification,” in Proceedings of the 26th International Joint Conference on Artificial Intelligence, 2017, pp. 4068–4074, doi: 10.5555/3171837.3171854.D. Ma and others, “Interactive attention networks for aspect-level sentiment classification,” in Proceedings of the 26th International Joint Conference on Artificial Intelligence, 2017, pp. 4068–4074, doi: 10.5555/3171837.3171854.A. Y. Poletaev, I. V. Paramonov, and E. I. Boychuk, “Semantic rule-based sentiment detection algorithm for Russian publicism sentences,” Modeling and Analysis of Information Systems, vol. 30, no. 4, pp. 394–417, 2023, doi: 10.18255/1818-1015-2023-4-394-417.A. Y. Poletaev, I. V. Paramonov, and E. I. Boychuk, “Semantic rule-based sentiment detection algorithm for Russian publicism sentences,” Modeling and Analysis of Information Systems, vol. 30, no. 4, pp. 394–417, 2023, doi: 10.18255/1818-1015-2023-4-394-417.A. Y. Poletaev, I. V. Paramonov, and E. I. Boychuk, “Algorithm of constituency tree from dependency tree construction for a Russian-language sentence,” Informatics and Automation, vol. 22, no. 6, pp. 1323–1353, 2023, doi: 10.15622/ia.22.6.3.A. Y. Poletaev, I. V. Paramonov, and E. I. Boychuk, “Algorithm of constituency tree from dependency tree construction for a Russian-language sentence,” Informatics and Automation, vol. 22, no. 6, pp. 1323–1353, 2023, doi: 10.15622/ia.22.6.3.D. Chandrasekaran and V. Mago, “Evolution of semantic similarity — a survey,” ACM Computing Surveys (CSUR), vol. 54, no. 2, pp. 1–37, 2021, doi: 10.1145/3440755.D. Chandrasekaran and V. Mago, “Evolution of semantic similarity — a survey,” ACM Computing Surveys (CSUR), vol. 54, no. 2, pp. 1–37, 2021, doi: 10.1145/3440755.A. Kutuzov and E. Kuzmenko, “WebVectors: A Toolkit for Building Web Interfaces for Vector Semantic Models,” in Analysis of Images, Social Networks and Texts: 5th International Conference, AIST 2016, Yekaterinburg, Russia, April 7--9, 2016, Revised Selected Papers, Cham: Springer International Publishing, 2017, pp. 155–161.A. Kutuzov and E. Kuzmenko, “WebVectors: A Toolkit for Building Web Interfaces for Vector Semantic Models,” in Analysis of Images, Social Networks and Texts: 5th International Conference, AIST 2016, Yekaterinburg, Russia, April 7--9, 2016, Revised Selected Papers, Cham: Springer International Publishing, 2017, pp. 155–161.M. Korobov, “Morphological Analyzer and Generator for Russian and Ukrainian Languages,” in Analysis of Images, Social Networks and Texts, vol. 542, Springer International Publishing, 2015, pp. 320–332.M. Korobov, “Morphological Analyzer and Generator for Russian and Ukrainian Languages,” in Analysis of Images, Social Networks and Texts, vol. 542, Springer International Publishing, 2015, pp. 320–332.P. Qi and others, “Stanza: A Python Natural Language Processing Toolkit for Many Human Languages,” in Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics: System Demonstrations, 2020, pp. 101–108, doi: 10.18653/v1/2020.acl-demos.14.P. Qi and others, “Stanza: A Python Natural Language Processing Toolkit for Many Human Languages,” in Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics: System Demonstrations, 2020, pp. 101–108, doi: 10.18653/v1/2020.acl-demos.14.

The authors declare that there are no conflicts of interest present.