<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE article PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Publishing DTD v1.3 20210610//EN" "JATS-journalpublishing1-3.dtd">
<article article-type="research-article" dtd-version="1.3" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xml:lang="ru"><front><journal-meta><journal-id journal-id-type="publisher-id">mais</journal-id><journal-title-group><journal-title xml:lang="ru">Моделирование и анализ информационных систем</journal-title><trans-title-group xml:lang="en"><trans-title>Modeling and Analysis of Information Systems</trans-title></trans-title-group></journal-title-group><issn pub-type="ppub">1818-1015</issn><issn pub-type="epub">2313-5417</issn><publisher><publisher-name>Yaroslavl State University</publisher-name></publisher></journal-meta><article-meta><article-id pub-id-type="doi">10.18255/1818-1015-2025-1-80-94</article-id><article-id custom-type="elpub" pub-id-type="custom">mais-1917</article-id><article-categories><subj-group subj-group-type="heading"><subject>Research Article</subject></subj-group><subj-group subj-group-type="section-heading" xml:lang="ru"><subject>Artificial Intelligence</subject></subj-group></article-categories><title-group><article-title>Иерархическая классификация научных статей при помощи глубокого обучения (на примере иерархии УДК)</article-title><trans-title-group xml:lang="en"><trans-title>Hierarchical classification of scientific articles using deep learning (using the UDC hierarchy as an example)</trans-title></trans-title-group></title-group><contrib-group><contrib contrib-type="author" corresp="yes"><contrib-id contrib-id-type="orcid">https://orcid.org/0009-0004-4154-5522</contrib-id><name-alternatives><name name-style="eastern" xml:lang="ru"><surname>Мамедов</surname><given-names>Валентин Юрьевич</given-names></name><name name-style="western" xml:lang="en"><surname>Mamedov</surname><given-names>Valentin Y.</given-names></name></name-alternatives><email xlink:type="simple">v.mamedov@g.nsu.ru</email><xref ref-type="aff" rid="aff-1"/></contrib><contrib contrib-type="author" corresp="yes"><contrib-id contrib-id-type="orcid">https://orcid.org/0009-0002-8484-7366</contrib-id><name-alternatives><name name-style="eastern" xml:lang="ru"><surname>Ковалевский</surname><given-names>Данил Анатольевич</given-names></name><name name-style="western" xml:lang="en"><surname>Kovalevsky</surname><given-names>Danil A.</given-names></name></name-alternatives><email xlink:type="simple">d.kovalevskii@g.nsu.ru</email><xref ref-type="aff" rid="aff-1"/></contrib><contrib contrib-type="author" corresp="yes"><contrib-id contrib-id-type="orcid">https://orcid.org/0000-0003-4464-1355</contrib-id><name-alternatives><name name-style="eastern" xml:lang="ru"><surname>Морозов</surname><given-names>Дмитрий Алексеевич</given-names></name><name name-style="western" xml:lang="en"><surname>Morozov</surname><given-names>Dmitry A.</given-names></name></name-alternatives><email xlink:type="simple">morozowdm@gmail.com</email><xref ref-type="aff" rid="aff-1"/></contrib><contrib contrib-type="author" corresp="yes"><contrib-id contrib-id-type="orcid">https://orcid.org/0009-0005-7651-6948</contrib-id><name-alternatives><name name-style="eastern" xml:lang="ru"><surname>Столяров</surname><given-names>Степан Сергеевич</given-names></name><name name-style="western" xml:lang="en"><surname>Stolyarov</surname><given-names>Stepan S.</given-names></name></name-alternatives><email xlink:type="simple">s.stolyarov@g.nsu.ru</email><xref ref-type="aff" rid="aff-1"/></contrib><contrib contrib-type="author" corresp="yes"><contrib-id contrib-id-type="orcid">https://orcid.org/0000-0001-9912-6364</contrib-id><name-alternatives><name name-style="eastern" xml:lang="ru"><surname>Оспичев</surname><given-names>Сергей Сергеевич</given-names></name><name name-style="western" xml:lang="en"><surname>Ospichev</surname><given-names>Sergey S.</given-names></name></name-alternatives><email xlink:type="simple">s.ospichev@nsu.ru</email><xref ref-type="aff" rid="aff-1"/></contrib></contrib-group><aff-alternatives id="aff-1"><aff xml:lang="ru"><institution>Новосибирский национальный исследовательский государственный университет</institution><country>Россия</country></aff><aff xml:lang="en"><institution>Novosibirsk National Research State University</institution><country>Russian Federation</country></aff></aff-alternatives><pub-date pub-type="collection"><year>2025</year></pub-date><pub-date pub-type="epub"><day>22</day><month>03</month><year>2025</year></pub-date><volume>32</volume><issue>1</issue><fpage>80</fpage><lpage>94</lpage><permissions><copyright-statement>Copyright &amp;#x00A9; Мамедов В.Ю., Ковалевский Д.А., Морозов Д.А., Столяров С.С., Оспичев С.С., 2025</copyright-statement><copyright-year>2025</copyright-year><copyright-holder xml:lang="ru">Мамедов В.Ю., Ковалевский Д.А., Морозов Д.А., Столяров С.С., Оспичев С.С.</copyright-holder><copyright-holder xml:lang="en">Mamedov V.Y., Kovalevsky D.A., Morozov D.A., Stolyarov S.S., Ospichev S.S.</copyright-holder><license xml:lang="ru" license-type="creative-commons-attribution" xlink:href="https://creativecommons.org/licenses/by/4.0/" xlink:type="simple"><license-p>Данная работа распространяется под лицензией Creative Commons Attribution 4.0.</license-p></license><license xml:lang="en" license-type="creative-commons-attribution" xlink:href="https://creativecommons.org/licenses/by/4.0/" xlink:type="simple"><license-p>This work is licensed under a Creative Commons Attribution 4.0 License.</license-p></license></permissions><self-uri xlink:href="https://www.mais-journal.ru/jour/article/view/1917">https://www.mais-journal.ru/jour/article/view/1917</self-uri><abstract><p>В условиях стремительного роста числа научных публикаций актуальной задачей становится разработка эффективных инструментов для их систематизации и поиска. Одним из таких инструментов является универсальная десятичная классификация (УДК), которая позволяет структурировать статьи по тематическим областям. Однако ручное присвоение кодов УДК зачастую оказывается неточным или недостаточно детализированным, что снижает эффективность использования этого подхода. В данной статье предлагается подход к автоматическому присвоению кодов УДК научным статьям с использованием моделей на основе архитектуры BERT. Для обучения и оценки модели был использован набор данных, содержащий более 19 тысяч статей по математике и смежным наукам. Мы разработали две специализированные метрики качества, учитывающие иерархическую природу УДК: иерархическую классификационную точность и иерархическую рекомендательную точность. Кроме того, мы предложили несколько стратегий преобразования иерархических меток в плоские. В ходе экспериментов нам удалось достичь значения иерархической рекомендательной точности 0,8220. Дополнительно проведено слепое тестирование с участием экспертов, которое выявило, что часть расхождений между эталонными и сгенерированными метками обусловлена некорректным выбором кода УДК авторами статей. Предложенный подход демонстрирует высокий потенциал для автоматической классификации научных статей и может быть адаптирован для других иерархических систем классификации.</p></abstract><trans-abstract xml:lang="en"><p>The exponential growth in scientific publications has heightened the need for robust tools to organize and retrieve research effectively. The Universal Decimal Classification (UDC) serves as a valuable framework for categorizing articles by subject area. However, manual assignment of UDC codes is often prone to inaccuracies or oversimplification, limiting its utility. In this study, we present a novel approach for the automated assignment of UDC codes to scientific articles using BERT-based models. Our methodology was trained and evaluated on a dataset comprising over 19,000 articles in mathematics and related disciplines. To address the hierarchical structure of UDC, we developed two specialized evaluation metrics: hierarchical classification accuracy and hierarchical recommendation accuracy. We also explored multiple strategies for flattening hierarchical labels. Our results demonstrated a hierarchical recommendation accuracy of 0.8220. Furthermore, blind expert evaluation revealed that discrepancies between reference and predicted labels often stem from errors in the original UDC code assignments by article authors. Our approach demonstrates strong potential for automating the classification of scientific articles and can be extended to other hierarchical classification systems.</p></trans-abstract><kwd-group xml:lang="ru"><kwd>классификация текстов</kwd><kwd>иерархическая классификация текстов</kwd><kwd>универсальный десятичный классификатор</kwd><kwd>глубокое обучение</kwd></kwd-group><kwd-group xml:lang="en"><kwd>text classification</kwd><kwd>hierarchical text classification</kwd><kwd>universal decimal classifier</kwd><kwd>deep learning</kwd></kwd-group></article-meta></front><back><ref-list><title>References</title><ref id="cit1"><label>1</label><citation-alternatives><mixed-citation xml:lang="ru">M. Gusenbauer, “Google Scholar to overshadow them all? Comparing the sizes of 12 academic search engines and bibliographic databases,” Scientometrics, vol. 118, pp. 177–214, 2019, doi: 10.1007/s11192-018-2958-5.</mixed-citation><mixed-citation xml:lang="en">M. Gusenbauer, “Google Scholar to overshadow them all? Comparing the sizes of 12 academic search engines and bibliographic databases,” Scientometrics, vol. 118, pp. 177–214, 2019, doi: 10.1007/s11192-018-2958-5.</mixed-citation></citation-alternatives></ref><ref id="cit2"><label>2</label><citation-alternatives><mixed-citation xml:lang="ru">M. Fire and C. Guestrin, “Over-optimization of academic publishing metrics: Observing Goodhart's Law in action,” GigaScience, vol. 8, p. giz053, Jun. 2019, doi: 10.1093/gigascience/giz053.</mixed-citation><mixed-citation xml:lang="en">M. Fire and C. Guestrin, “Over-optimization of academic publishing metrics: Observing Goodhart's Law in action,” GigaScience, vol. 8, p. giz053, Jun. 2019, doi: 10.1093/gigascience/giz053.</mixed-citation></citation-alternatives></ref><ref id="cit3"><label>3</label><citation-alternatives><mixed-citation xml:lang="ru">R. Martinez-Cruz, A. J. Lopez-Lopez, and J. Portela, “ChatGPT vs state-of-the-art models: a benchmarking study in keyphrase generation task,” Applied Intelligence, vol. 55, no. 1, pp. 1–25, 2025, doi: 10.1007/s10489-024-05901-4.</mixed-citation><mixed-citation xml:lang="en">R. Martinez-Cruz, A. J. Lopez-Lopez, and J. Portela, “ChatGPT vs state-of-the-art models: a benchmarking study in keyphrase generation task,” Applied Intelligence, vol. 55, no. 1, pp. 1–25, 2025, doi: 10.1007/s10489-024-05901-4.</mixed-citation></citation-alternatives></ref><ref id="cit4"><label>4</label><citation-alternatives><mixed-citation xml:lang="ru">M. Song et al., “Is ChatGPT A Good Keyphrase Generator? A Preliminary Study.” 2023.</mixed-citation><mixed-citation xml:lang="en">M. Song et al., “Is ChatGPT A Good Keyphrase Generator? A Preliminary Study.” 2023.</mixed-citation></citation-alternatives></ref><ref id="cit5"><label>5</label><citation-alternatives><mixed-citation xml:lang="ru">A. Glazkova, D. Morozov, and T. Garipov, “Key Algorithms for Keyphrase Generation: Instruction-Based LLMs for Russian Scientific Keyphrases.” 2024.</mixed-citation><mixed-citation xml:lang="en">A. Glazkova, D. Morozov, and T. Garipov, “Key Algorithms for Keyphrase Generation: Instruction-Based LLMs for Russian Scientific Keyphrases.” 2024.</mixed-citation></citation-alternatives></ref><ref id="cit6"><label>6</label><citation-alternatives><mixed-citation xml:lang="ru">A. V. Glazkova, D. A. Morozov, M. S. Vorobeva, and A. A. Stupnikov, “Keyword Generation for Russian-Language Scientific Texts Using the mT5 Model,” Automatic Control and Computer Sciences, vol. 58, no. 7, pp. 995–1002, 2024, doi: 10.3103/S014641162470041X.</mixed-citation><mixed-citation xml:lang="en">A. V. Glazkova, D. A. Morozov, M. S. Vorobeva, and A. A. Stupnikov, “Keyword Generation for Russian-Language Scientific Texts Using the mT5 Model,” Automatic Control and Computer Sciences, vol. 58, no. 7, pp. 995–1002, 2024, doi: 10.3103/S014641162470041X.</mixed-citation></citation-alternatives></ref><ref id="cit7"><label>7</label><citation-alternatives><mixed-citation xml:lang="ru">K. Kowsari, K. Jafari Meimandi, M. Heidarysafa, S. Mendu, L. Barnes, and D. Brown, “Text Classification Algorithms: A Survey,” Information, vol. 10, no. 4, p. 150, 2019, doi: 10.3390/info10040150.</mixed-citation><mixed-citation xml:lang="en">K. Kowsari, K. Jafari Meimandi, M. Heidarysafa, S. Mendu, L. Barnes, and D. Brown, “Text Classification Algorithms: A Survey,” Information, vol. 10, no. 4, p. 150, 2019, doi: 10.3390/info10040150.</mixed-citation></citation-alternatives></ref><ref id="cit8"><label>8</label><citation-alternatives><mixed-citation xml:lang="ru">S. Minaee, N. Kalchbrenner, E. Cambria, N. Nikzad, M. Chenaghlu, and J. Gao, “Deep Learning--based Text Classification: A Comprehensive Review,” ACM Computing Surveys, vol. 54, no. 3, 2021, doi: 10.1145/3439726.</mixed-citation><mixed-citation xml:lang="en">S. Minaee, N. Kalchbrenner, E. Cambria, N. Nikzad, M. Chenaghlu, and J. Gao, “Deep Learning--based Text Classification: A Comprehensive Review,” ACM Computing Surveys, vol. 54, no. 3, 2021, doi: 10.1145/3439726.</mixed-citation></citation-alternatives></ref><ref id="cit9"><label>9</label><citation-alternatives><mixed-citation xml:lang="ru">S. Garg and G. Ramakrishnan, “BAE: BERT-based Adversarial Examples for Text Classification,” in Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing, 2020, pp. 6174–6181, doi: 10.18653/v1/2020.emnlp-main.498.</mixed-citation><mixed-citation xml:lang="en">S. Garg and G. Ramakrishnan, “BAE: BERT-based Adversarial Examples for Text Classification,” in Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing, 2020, pp. 6174–6181, doi: 10.18653/v1/2020.emnlp-main.498.</mixed-citation></citation-alternatives></ref><ref id="cit10"><label>10</label><citation-alternatives><mixed-citation xml:lang="ru">X. Sun et al., “Text Classification via Large Language Models,” in Findings of the Association for Computational Linguistics, 2023, pp. 8990–9005, doi: 10.18653/v1/2023.findings-emnlp.603.</mixed-citation><mixed-citation xml:lang="en">X. Sun et al., “Text Classification via Large Language Models,” in Findings of the Association for Computational Linguistics, 2023, pp. 8990–9005, doi: 10.18653/v1/2023.findings-emnlp.603.</mixed-citation></citation-alternatives></ref><ref id="cit11"><label>11</label><citation-alternatives><mixed-citation xml:lang="ru">K. Kowsari, D. E. Brown, M. Heidarysafa, K. Jafari Meimandi, M. S. Gerber, and L. E. Barnes, “HDLTex: Hierarchical Deep Learning for Text Classification,” in Proceedings of the 16th IEEE International Conference on Machine Learning and Applications, 2017, pp. 364–371, doi: 10.1109/ICMLA.2017.0-134.</mixed-citation><mixed-citation xml:lang="en">K. Kowsari, D. E. Brown, M. Heidarysafa, K. Jafari Meimandi, M. S. Gerber, and L. E. Barnes, “HDLTex: Hierarchical Deep Learning for Text Classification,” in Proceedings of the 16th IEEE International Conference on Machine Learning and Applications, 2017, pp. 364–371, doi: 10.1109/ICMLA.2017.0-134.</mixed-citation></citation-alternatives></ref><ref id="cit12"><label>12</label><citation-alternatives><mixed-citation xml:lang="ru">S. Strydom, A. M. Dreyer, and B. van der Merwe, “Automatic assignment of diagnosis codes to free-form text medical note,” Journal of Universal Computer Science, vol. 29, no. 4, pp. 349–373, 2023, doi: 10.3897/jucs.89923.</mixed-citation><mixed-citation xml:lang="en">S. Strydom, A. M. Dreyer, and B. van der Merwe, “Automatic assignment of diagnosis codes to free-form text medical note,” Journal of Universal Computer Science, vol. 29, no. 4, pp. 349–373, 2023, doi: 10.3897/jucs.89923.</mixed-citation></citation-alternatives></ref><ref id="cit13"><label>13</label><citation-alternatives><mixed-citation xml:lang="ru">R. A. Stein, P. A. Jaques, and J. F. Valiati, “An analysis of hierarchical text classification using word embeddings,” Information Sciences, vol. 471, pp. 216–232, 2019, doi: 10.1016/j.ins.2018.09.001.</mixed-citation><mixed-citation xml:lang="en">R. A. Stein, P. A. Jaques, and J. F. Valiati, “An analysis of hierarchical text classification using word embeddings,” Information Sciences, vol. 471, pp. 216–232, 2019, doi: 10.1016/j.ins.2018.09.001.</mixed-citation></citation-alternatives></ref><ref id="cit14"><label>14</label><citation-alternatives><mixed-citation xml:lang="ru">D. D. Lewis, Y. Yang, T. G. Rose, and F. Li, “RCV1: A New Benchmark Collection for Text Categorization Research,” Journal of Machine Learning Research, vol. 5, pp. 361–397, 2004.</mixed-citation><mixed-citation xml:lang="en">D. D. Lewis, Y. Yang, T. G. Rose, and F. Li, “RCV1: A New Benchmark Collection for Text Categorization Research,” Journal of Machine Learning Research, vol. 5, pp. 361–397, 2004.</mixed-citation></citation-alternatives></ref><ref id="cit15"><label>15</label><citation-alternatives><mixed-citation xml:lang="ru">Y. Wang et al., “Towards Better Hierarchical Text Classification with Data Generation,” in Findings of the Association for Computational Linguistics, 2023, pp. 7722–7739, doi: 10.18653/v1/2023.findings-acl.489.</mixed-citation><mixed-citation xml:lang="en">Y. Wang et al., “Towards Better Hierarchical Text Classification with Data Generation,” in Findings of the Association for Computational Linguistics, 2023, pp. 7722–7739, doi: 10.18653/v1/2023.findings-acl.489.</mixed-citation></citation-alternatives></ref><ref id="cit16"><label>16</label><citation-alternatives><mixed-citation xml:lang="ru">A. Zangari, M. Marcuzzo, M. Rizzo, L. Giudice, A. Albarelli, and A. Gasparetto, “Hierarchical Text Classification and Its Foundations: A Review of Current Research,” Electronics, vol. 13, no. 7, p. 1199, 2024, doi: 10.3390/electronics13071199.</mixed-citation><mixed-citation xml:lang="en">A. Zangari, M. Marcuzzo, M. Rizzo, L. Giudice, A. Albarelli, and A. Gasparetto, “Hierarchical Text Classification and Its Foundations: A Review of Current Research,” Electronics, vol. 13, no. 7, p. 1199, 2024, doi: 10.3390/electronics13071199.</mixed-citation></citation-alternatives></ref><ref id="cit17"><label>17</label><citation-alternatives><mixed-citation xml:lang="ru">M. Kragelj and M. Borstnar, “Automatic classification of older electronic texts into the Universal Decimal Classification-UDC,” Journal of Documentation, vol. 77, no. 3, pp. 755–776, 2021, doi: 10.1108/JD-06-2020-0092.</mixed-citation><mixed-citation xml:lang="en">M. Kragelj and M. Borstnar, “Automatic classification of older electronic texts into the Universal Decimal Classification-UDC,” Journal of Documentation, vol. 77, no. 3, pp. 755–776, 2021, doi: 10.1108/JD-06-2020-0092.</mixed-citation></citation-alternatives></ref><ref id="cit18"><label>18</label><citation-alternatives><mixed-citation xml:lang="ru">A. Y. Romanov, K. E. Lomotin, E. S. Kozlova, and A. L. Kolesnichenko, “Research of neural networks application efficiency in automatic scientific articles classification according to UDC,” in Proceedings of the International Siberian Conference on Control and Communications (SIBCON), 2016, pp. 1–5.</mixed-citation><mixed-citation xml:lang="en">A. Y. Romanov, K. E. Lomotin, E. S. Kozlova, and A. L. Kolesnichenko, “Research of neural networks application efficiency in automatic scientific articles classification according to UDC,” in Proceedings of the International Siberian Conference on Control and Communications (SIBCON), 2016, pp. 1–5.</mixed-citation></citation-alternatives></ref><ref id="cit19"><label>19</label><citation-alternatives><mixed-citation xml:lang="ru">O. Nevzorova and D. Almukhametov, “Towards a Recommender System for the Choice of UDC Code for Mathematical Articles,” in Supplementary Proceedings of the XXIII International Conference on Data Analytics and Management in Data Intensive Domains, 2021, pp. 54–62.</mixed-citation><mixed-citation xml:lang="en">O. Nevzorova and D. Almukhametov, “Towards a Recommender System for the Choice of UDC Code for Mathematical Articles,” in Supplementary Proceedings of the XXIII International Conference on Data Analytics and Management in Data Intensive Domains, 2021, pp. 54–62.</mixed-citation></citation-alternatives></ref><ref id="cit20"><label>20</label><citation-alternatives><mixed-citation xml:lang="ru">L. Wang, N. Yang, X. Huang, L. Yang, R. Majumder, and F. Wei, “Multilingual E5 Text Embeddings: A Technical Report.” 2024.</mixed-citation><mixed-citation xml:lang="en">L. Wang, N. Yang, X. Huang, L. Yang, R. Majumder, and F. Wei, “Multilingual E5 Text Embeddings: A Technical Report.” 2024.</mixed-citation></citation-alternatives></ref><ref id="cit21"><label>21</label><citation-alternatives><mixed-citation xml:lang="ru">A. Snegirev, M. Tikhonova, A. Maksimova, A. Fenogenova, and A. Abramov, “The Russian-focused embedders' exploration: ruMTEB benchmark and Russian embedding model design.” 2025.</mixed-citation><mixed-citation xml:lang="en">A. Snegirev, M. Tikhonova, A. Maksimova, A. Fenogenova, and A. Abramov, “The Russian-focused embedders' exploration: ruMTEB benchmark and Russian embedding model design.” 2025.</mixed-citation></citation-alternatives></ref><ref id="cit22"><label>22</label><citation-alternatives><mixed-citation xml:lang="ru">D. P. Kingma and J. Ba, “Adam: A Method for Stochastic Optimization.” 2017, [Online]. Available: https://arxiv.org/abs/1412.6980.</mixed-citation><mixed-citation xml:lang="en">D. P. Kingma and J. Ba, “Adam: A Method for Stochastic Optimization.” 2017, [Online]. Available: https://arxiv.org/abs/1412.6980.</mixed-citation></citation-alternatives></ref></ref-list><fn-group><fn fn-type="conflict"><p>The authors declare that there are no conflicts of interest present.</p></fn></fn-group></back></article>
