maisМоделирование и анализ информационных системModeling and Analysis of Information Systems1818-10152313-5417Yaroslavl State University10.18255/1818-1015-2022-4-334-347mais-1750Research ArticleTheory of DataКлассификация русскоязычных текстов по жанрам на основе современных эмбеддингов и ритмаClassification of Russian Texts by Genres Based on Modern Embeddings and Rhythmhttps://orcid.org/0000-0002-1742-3240ЛагутинаКсения ВладимировнаLagutinaKsenia Vladimirovnalagutinakv@mail.ruЯрославский государственный университет им. П. Г. ДемидоваРоссияP. G. Demidov Yaroslavl State UniversityRussian Federation202218122022294334347Copyright © Лагутина К.В., 20222022Лагутина К.В.Lagutina K.V.This work is licensed under a Creative Commons Attribution 4.0 License.https://www.mais-journal.ru/jour/article/view/1750

В статье исследуются современные векторные модели текстов для решения задачи классификации русскоязычных текстов по жанрам. Модели включают эмбеддинги ELMo, языковую модель BERT с предобучением и комплекс числовых ритмических характеристик на основе лексико-грамматических средств. Эксперименты проводились на корпусе из 10 000 текстов пяти жанров: романы, научные статьи, отзывы, посты из социальной сети Вконтакте, новости из OpenCorpora. Визуализация и анализ статистики для ритмических характеристик позволили выделить как наиболее разнообразные по ритму жанры: романы и отзывы, так и наименее - научные статьи. Именно эти жанры были впоследствии классифицированы лучше всего с помощью ритма и нейросети-классификатора LSTM. Кластеризация и классификация текстов по жанрам с помощью эмбеддингов ELMo и BERT позволила отделить один жанр от другого с небольшим количеством ошибок. F-мера мультиклассификации достигла 99%. Исследование подтверждает эффективность современных эмбеддингов в задачах компьютерной лингвистики, а также позволяет выделить достоинства и ограничения комплекса ритмических характеристик на материале классификации по жанрам.

The article investigates modern vector text models for solving the problem of genre classification of Russian-language texts. Models include ELMo embeddings, BERT language model with pre-training and a complex of numerical rhythm features based on lexico-grammatical features. The experiments were carried out on a corpus of 10,000 texts in five genres: novels, scientific articles, reviews, posts from the social network Vkontakte, news from OpenCorpora. Visualization and analysis of statistics for rhythm features made it possible to identify both the most diverse genres in terms of rhythm: novels and reviews, and the least ones: scientific articles. Subsequently, these genres were classified best with the help of rhythm features and the neural network-classifier LSTM. Clustering and classifying texts by genre using ELMo and BERT embeddings made it possible to separate one genre from another with a small number of errors. The multiclassification F-score reached 99%. The study confirms the efficiency of modern embeddings in the tasks of computational linguistics, and also allows to highlight the advantages and limitations of the complex of rhythm features on the material of genre classification.

стилометрияобработка естественного языкаритмические характеристикижанрыклассификация текстовBERTELMostylometrynatural language processingrhythm featuresgenrestext classificationBERTELMoReferencesL. A. Kochetova and V. V. Popov, "Research of Axiological Dominants in Press Release Genre based on Automatic Extraction of Key Words from Corpus”, Nauchnyi dialog, no. 6, 2019, In Russian.L. A. Kochetova and V. V. Popov, "Research of Axiological Dominants in Press Release Genre based on Automatic Extraction of Key Words from Corpus”, Nauchnyi dialog, no. 6, 2019, In Russian.B. Kessler, G. Numberg, and H. Schutze, "Automatic detection of text genre”, in Proceedings of the 35th Annual Meeting of the Association for Computational Linguistics and Eighth Conference of the European Chapter of the Association for Computational Linguistics, 1997, pp. 32-38.B. Kessler, G. Numberg, and H. Schutze, "Automatic detection of text genre”, in Proceedings of the 35th Annual Meeting of the Association for Computational Linguistics and Eighth Conference of the European Chapter of the Association for Computational Linguistics, 1997, pp. 32-38.A. Onan, "An ensemble scheme based on language function analysis and feature engineering for text genre classification”, Journal of Information Science, vol. 44, no. 1, pp. 28-47, 2018.A. Onan, "An ensemble scheme based on language function analysis and feature engineering for text genre classification”, Journal of Information Science, vol. 44, no. 1, pp. 28-47, 2018.Z. Dai and R. Huang, "A Joint Model for Structure-based News Genre Classification with Application to Text Summarization”, in Findings of the Association for Computational Linguistics:ACL-IJCNLP 2021, 2021, pp. 3332-3342.Z. Dai and R. Huang, "A Joint Model for Structure-based News Genre Classification with Application to Text Summarization”, in Findings of the Association for Computational Linguistics:ACL-IJCNLP 2021, 2021, pp. 3332-3342.K. V. Lagutina, N. S. Lagutina, and E. I. Boychuk, "Text classification by genre based on rhythm features”, Modeling and analysis of information systems, vol. 28, no. 3, pp. 280-291, 2021.K. V. Lagutina, N. S. Lagutina, and E. I. Boychuk, "Text classification by genre based on rhythm features”, Modeling and analysis of information systems, vol. 28, no. 3, pp. 280-291, 2021.K. Lagutina, A. Poletaev, N. Lagutina, E. Boychuk, and I. Paramonov, "Automatic extraction of rhythm figures and analysis of their dynamics in prose of 19th-21st centuries”, Proceedings of the 26th Conference of Open Innovations Association FRUCT, pp. 247-255, 2020.K. Lagutina, A. Poletaev, N. Lagutina, E. Boychuk, and I. Paramonov, "Automatic extraction of rhythm figures and analysis of their dynamics in prose of 19th-21st centuries”, Proceedings of the 26th Conference of Open Innovations Association FRUCT, pp. 247-255, 2020.M. Peters, M. Neumann, M. Iyyer, M. Gardner, C. Clark, K. Lee, and L. Zettlemoyer, "Deep Contextualized Word Representations”, in Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Papers), 2018, pp. 2227-2237.M. Peters, M. Neumann, M. Iyyer, M. Gardner, C. Clark, K. Lee, and L. Zettlemoyer, "Deep Contextualized Word Representations”, in Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Papers), 2018, pp. 2227-2237.J. Devlin, M.-W. Chang, K. Lee, and K. Toutanova, "BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding”, in Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers), 2019, pp. 4171-4186.J. Devlin, M.-W. Chang, K. Lee, and K. Toutanova, "BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding”, in Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers), 2019, pp. 4171-4186.C. Wang, P. Nulty, and D. Lillis, "A comparative study on word embeddings in deep learning for text classification”, in Proceedings of the 4th International Conference on Natural Language Processing and Information Retrieval, 2020, pp. 37-46.C. Wang, P. Nulty, and D. Lillis, "A comparative study on word embeddings in deep learning for text classification”, in Proceedings of the 4th International Conference on Natural Language Processing and Information Retrieval, 2020, pp. 37-46.Y. Kuratov and M. Arkhipov, "Adaptation of deep bidirectional multilingual transformers for Russian language”, in Komp’juternaja Lingvistika i Intellektual’nye Tehnologii, 2019, pp. 333-339.Y. Kuratov and M. Arkhipov, "Adaptation of deep bidirectional multilingual transformers for Russian language”, in Komp’juternaja Lingvistika i Intellektual’nye Tehnologii, 2019, pp. 333-339.A. Kutuzov, L. Pivovarova, etal., "RuShiftEval: a shared task on semantic shift detection for Russian”, in Computational Linguistics and Intellectual Technologies Papers from the Annual International Conference “Dialogue” (2021), vol. 20, 2021, pp. 533-545.A. Kutuzov, L. Pivovarova, etal., "RuShiftEval: a shared task on semantic shift detection for Russian”, in Computational Linguistics and Intellectual Technologies Papers from the Annual International Conference “Dialogue” (2021), vol. 20, 2021, pp. 533-545.J. Rodina, Y. Trofimova, A. Kutuzov, and E. Artemova, "ELMo and BERT in semantic change detection for Russian”, in International Conference on Analysis of Images, Social Networks and Texts, Springer, 2020, pp. 175-186.J. Rodina, Y. Trofimova, A. Kutuzov, and E. Artemova, "ELMo and BERT in semantic change detection for Russian”, in International Conference on Analysis of Images, Social Networks and Texts, Springer, 2020, pp. 175-186.A. V. Glazkova, "Topical classification of text fragments accounting for their nearest context”, Automation and Remote Control, vol. 81, no. 12, pp. 2262-2276, 2020.A. V. Glazkova, "Topical classification of text fragments accounting for their nearest context”, Automation and Remote Control, vol. 81, no. 12, pp. 2262-2276, 2020.I. A. Batraeva, A. D. Nartsev, and A. S. Lezgyan, "Using the analysis of semantic proximity of words in solving the problem of determining the genre of texts within deep learning”, Vestnik Tomskogo gosudarstvennogo universiteta. Upravlenie vychislitelnaja tehnika i informatika, no. 50, pp. 14-22, 2020, In Russian.I. A. Batraeva, A. D. Nartsev, and A. S. Lezgyan, "Using the analysis of semantic proximity of words in solving the problem of determining the genre of texts within deep learning”, Vestnik Tomskogo gosudarstvennogo universiteta. Upravlenie vychislitelnaja tehnika i informatika, no. 50, pp. 14-22, 2020, In Russian.V. Bocharov, S. Alexeeva, D. Granovsky, E. Protopopova, M. Stepanova, and A. Surikov, "Crowdsourcing morphological annotation”, in Computational Linguistics and Intellectual Technologies: Papers from the Annual International Conference ’’Dialogue”. Volume 1, 2013, pp. 109-114.V. Bocharov, S. Alexeeva, D. Granovsky, E. Protopopova, M. Stepanova, and A. Surikov, "Crowdsourcing morphological annotation”, in Computational Linguistics and Intellectual Technologies: Papers from the Annual International Conference ’’Dialogue”. Volume 1, 2013, pp. 109-114.K. Lagutina, N. Lagutina, E. Boychuk, V. Larionov, and I. Paramonov, "Authorship verification of literary texts with rhythm features”, in 28th Conference of Open Innovations Association FRUCT, IEEE, 2021, pp. 240-251.K. Lagutina, N. Lagutina, E. Boychuk, V. Larionov, and I. Paramonov, "Authorship verification of literary texts with rhythm features”, in 28th Conference of Open Innovations Association FRUCT, IEEE, 2021, pp. 240-251.

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