<?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">ntv</journal-id><journal-title-group><journal-title xml:lang="ru">Научно-технический вестник информационных технологий, механики и оптики</journal-title><trans-title-group xml:lang="en"><trans-title>Scientific and Technical Journal of Information Technologies, Mechanics and Optics</trans-title></trans-title-group></journal-title-group><issn pub-type="ppub">2226-1494</issn><issn pub-type="epub">2500-0373</issn><publisher><publisher-name>Университет ИТМО</publisher-name></publisher></journal-meta><article-meta><article-id pub-id-type="doi">10.17586/2226-1494-2026-26-2-331-336</article-id><article-id custom-type="elpub" pub-id-type="custom">ntv-595</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>КОМПЬЮТЕРНЫЕ СИСТЕМЫ И ИНФОРМАЦИОННЫЕ ТЕХНОЛОГИИ</subject></subj-group><subj-group subj-group-type="section-heading" xml:lang="en"><subject>COMPUTER SCIENCE</subject></subj-group></article-categories><title-group><article-title>Предсказание связей в эго-графах с GNN</article-title><trans-title-group xml:lang="en"><trans-title>Ego-net link prediction with GNN</trans-title></trans-title-group></title-group><contrib-group><contrib contrib-type="author" corresp="yes"><contrib-id contrib-id-type="orcid">https://orcid.org/0009-0002-7334-2856</contrib-id><name-alternatives><name name-style="eastern" xml:lang="ru"><surname>Замятин</surname><given-names>Е. И.</given-names></name><name name-style="western" xml:lang="en"><surname>Zamyatin</surname><given-names>E. I.</given-names></name></name-alternatives><bio xml:lang="ru"><p>Замятин Евгений Игоревич — аспирант</p><p>Санкт-Петербург, 197101</p><p>sc 57207471477</p></bio><bio xml:lang="en"><p>Evgeny I. Zamyatin — PhD Student</p><p>Saint Petersburg, 197101</p><p>sc 57207471477</p></bio><email xlink:type="simple">ezamyatin@itmo.ru</email><xref ref-type="aff" rid="aff-1"/></contrib></contrib-group><aff-alternatives id="aff-1"><aff xml:lang="ru"><institution>Университет ИТМО</institution><country>Россия</country></aff><aff xml:lang="en"><institution>ITMO University</institution><country>Russian Federation</country></aff></aff-alternatives><pub-date pub-type="collection"><year>2026</year></pub-date><pub-date pub-type="epub"><day>20</day><month>04</month><year>2026</year></pub-date><volume>26</volume><issue>2</issue><fpage>331</fpage><lpage>336</lpage><permissions><copyright-statement>Copyright &amp;#x00A9; Замятин Е.И., 2026</copyright-statement><copyright-year>2026</copyright-year><copyright-holder xml:lang="ru">Замятин Е.И.</copyright-holder><copyright-holder xml:lang="en">Zamyatin E.I.</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://ntv.elpub.ru/jour/article/view/595">https://ntv.elpub.ru/jour/article/view/595</self-uri><abstract><p>Введение. Задача предсказания связей в графе — одна из ключевых задач в области анализа социальных сетей. В основе одного из распространенных способов построения таких систем лежит идея декомпозиции задачи на два уровня. На первом уровне формируются предсказания возникновения связей внутри эго-графов, на втором — агрегация результатов и формирование итоговой выдачи. Точность таких систем определяется моделью первого уровня. Обычно здесь используют эвристические методы. Основное внимание в данной работе уделено разработке и исследованию новой модели с учителем для улучшения качества предсказаний связей внутри эго-графов. Неоднородность свойств ребер, отсутствие признаков вершин, а также динамическая природа эго-графов выделяют эту задачу среди остальных. Метод. Предлагаемый метод относится к классу графовых нейронных сетей. Его отличительная особенность в способности эффективно учитывать топологию графа вместе со свойствами ребер, при этом не опираясь на признаки вершин. Такой эффект удается достичь за счет моделирования скрытого состояния именно пар вершин, а не каждой вершины в отдельности. Итеративная сущность модели позволяет распространять знание о взаимосвязях вершин, с каждым шагом увеличивая сложность учитываемых структур. Основные результаты. Для замеров эффективности модели была использована база данных Ego-VK, состоящая из набора эго-графов подвыборки пользователей социальной сети «ВКонтакте». Проведено сравнение с классическим методом предсказания связей Adamic-Adar, а также с современными подходами на основе графовых нейронных сетей. Эксперименты показали, что предлагаемая модель значительно превосходит бейзлайны с точки зрения метрики качества ранжирования NDCG@5. Обсуждение. Полученные результаты свидетельствуют о высокой эффективности предложенной модели, а возможность интеграции в декомпозированные системы делает ее широко применимой в индустрии.</p></abstract><trans-abstract xml:lang="en"><p>The task of link prediction is one of the key challenges in the field of social network analysis. The common way to build such systems is based on the idea of decomposing a task into two levels. At the first level, links within ego-nets are predicted; at the second, the results are aggregated to form the final predictions. The accuracy of such systems depends on the first-level model. Heuristic methods are usually used here. The focus of this work is on developing a new supervised model to improve the quality of link prediction within ego-nets. The heterogeneity of the edge attributes, the absence of node features, and the dynamic nature of ego-nets distinguish this task from others. The proposed method belongs to the class of graph neural networks. Its key feature is the ability to effectively consider the topology of the graph along with the attributes of the edges, without relying on the properties of the nodes. This effect is achieved by modeling the hidden state of node pairs, rather than the state of each node individually. The iterative nature of the model makes it possible to propagate knowledge about the relationships between nodes, increasing the complexity of the structures considered with each step. To measure the accuracy of the model, the Ego-VK dataset was used. This dataset consists of a set of ego-nets from a subsample of users of the VKontakte social network. The model is compared with the classical Adamic-Adar method as well as modern approaches based on graph neural networks. Experiments show that the proposed model is significantly superior to the baselines with respect to NDCG@5 ranking quality metric. The results demonstrate the high effectiveness of the proposed model, and the possibility of integration into distributed systems makes it widely applicable in the industry.</p></trans-abstract><kwd-group xml:lang="ru"><kwd>информационный поиск</kwd><kwd>рекомендательные системы</kwd><kwd>графовые нейросети</kwd><kwd>анализ социальных графов</kwd><kwd>эго-графы</kwd></kwd-group><kwd-group xml:lang="en"><kwd>information retrieval</kwd><kwd>recommender systems</kwd><kwd>graph neural networks</kwd><kwd>social networks analysis</kwd><kwd>ego-nets</kwd></kwd-group><funding-group><funding-statement xml:lang="ru">Работа выполнена в рамках проекта № 625135 «Разработка программных модулей для эффективного обучения и применения моделей глубокого обучения».</funding-statement><funding-statement xml:lang="en">The research was carried out under the Project No. 625135 “Development of software modules for effective training and inference of deep learning models”.</funding-statement></funding-group></article-meta></front><back><ref-list><title>References</title><ref id="cit1"><label>1</label><citation-alternatives><mixed-citation xml:lang="ru">Keramatfar A., Rafiee M., Amirkhani H. Graph Neural Networks: a bibliometrics overview. Machine Learning with Applications, 2022, vol. 10, pp. 100401. https://doi.org/10.1016/j.mlwa.2022.100401</mixed-citation><mixed-citation xml:lang="en">Keramatfar A., Rafiee M., Amirkhani H. Graph Neural Networks: a bibliometrics overview. Machine Learning with Applications, 2022, vol. 10, pp. 100401. https://doi.org/10.1016/j.mlwa.2022.100401</mixed-citation></citation-alternatives></ref><ref id="cit2"><label>2</label><citation-alternatives><mixed-citation xml:lang="ru">Abadal S., Jain A., Guirado R., López-Alonso J., Alarcón E. Computing Graph Neural Networks: a survey from algorithms to accelerators. ACM Computing Surveys (CSUR), 2021, vol. 54, no. 9, pp. 1–38. https://doi.org/10.1145/3477141</mixed-citation><mixed-citation xml:lang="en">Abadal S., Jain A., Guirado R., López-Alonso J., Alarcón E. Computing Graph Neural Networks: a survey from algorithms to accelerators. ACM Computing Surveys (CSUR), 2021, vol. 54, no. 9, pp. 1–38. https://doi.org/10.1145/3477141</mixed-citation></citation-alternatives></ref><ref id="cit3"><label>3</label><citation-alternatives><mixed-citation xml:lang="ru">Ragunathan K., Selvarajah K., Kobti, Z. Link prediction by analyzing common neighbors based subgraphs using convolutional neural network. Frontiers in Artificial Intelligence and Applications, 2020, vol. 325, pp. 1906-1913. https://doi.org/10.3233/faia200308</mixed-citation><mixed-citation xml:lang="en">Ragunathan K., Selvarajah K., Kobti, Z. Link prediction by analyzing common neighbors based subgraphs using convolutional neural network. Frontiers in Artificial Intelligence and Applications, 2020, vol. 325, pp. 1906-1913. https://doi.org/10.3233/faia200308</mixed-citation></citation-alternatives></ref><ref id="cit4"><label>4</label><citation-alternatives><mixed-citation xml:lang="ru">Zhang M., Chen Y. Link prediction based on graph neural networks. Proc. of the 32nd International Conference on Neural Information Processing Systems, 2018, pp. 5171–5181.</mixed-citation><mixed-citation xml:lang="en">Zhang M., Chen Y. Link prediction based on graph neural networks. Proc. of the 32nd International Conference on Neural Information Processing Systems, 2018, pp. 5171–5181.</mixed-citation></citation-alternatives></ref><ref id="cit5"><label>5</label><citation-alternatives><mixed-citation xml:lang="ru">Epasto A., Lattanzi S., Mirrokni V., Sebe I.O., Taei A., Verma S. Egonet community mining applied to friend suggestion. Proceedings of the VLDB Endowment, 2015, vol. 9, no. 4, pp. 324–335. https://doi.org/10.14778/2856318.2856327</mixed-citation><mixed-citation xml:lang="en">Epasto A., Lattanzi S., Mirrokni V., Sebe I.O., Taei A., Verma S. Egonet community mining applied to friend suggestion. Proceedings of the VLDB Endowment, 2015, vol. 9, no. 4, pp. 324–335. https://doi.org/10.14778/2856318.2856327</mixed-citation></citation-alternatives></ref><ref id="cit6"><label>6</label><citation-alternatives><mixed-citation xml:lang="ru">Perozzi B., Al-Rfou R., Skiena S. Deepwalk: online learning of social representations. Proc. of the 20th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2014, pp. 701–710. https://doi.org/10.1145/2623330.2623732</mixed-citation><mixed-citation xml:lang="en">Perozzi B., Al-Rfou R., Skiena S. Deepwalk: online learning of social representations. Proc. of the 20th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2014, pp. 701–710. https://doi.org/10.1145/2623330.2623732</mixed-citation></citation-alternatives></ref><ref id="cit7"><label>7</label><citation-alternatives><mixed-citation xml:lang="ru">Zhu Z., Zhang Z., Xhonneux L.-P., Tang J. Neural bellman-ford networks: A general graph neural network framework for link prediction. Proc. of the 35th International Conference on Neural Information Processing, 2021, pp. 29476–29490.</mixed-citation><mixed-citation xml:lang="en">Zhu Z., Zhang Z., Xhonneux L.-P., Tang J. Neural bellman-ford networks: A general graph neural network framework for link prediction. Proc. of the 35th International Conference on Neural Information Processing, 2021, pp. 29476–29490.</mixed-citation></citation-alternatives></ref><ref id="cit8"><label>8</label><citation-alternatives><mixed-citation xml:lang="ru">Suri S., Vassilvitskii S. Counting triangles and the curse of the last reducer. Proc. of the 20th International Conference on World Wide Web, 2011, pp. 607–614. https://doi.org/10.1145/1963405.1963491</mixed-citation><mixed-citation xml:lang="en">Suri S., Vassilvitskii S. Counting triangles and the curse of the last reducer. Proc. of the 20th International Conference on World Wide Web, 2011, pp. 607–614. https://doi.org/10.1145/1963405.1963491</mixed-citation></citation-alternatives></ref><ref id="cit9"><label>9</label><citation-alternatives><mixed-citation xml:lang="ru">Schank T. Algorithmic Aspects of Triangle-Based Network Analysis: Ph.D. dissertation. Universitat Karlsruhe, Fakultat fur Informatik, 2007.</mixed-citation><mixed-citation xml:lang="en">Schank T. Algorithmic Aspects of Triangle-Based Network Analysis: Ph.D. dissertation. Universitat Karlsruhe, Fakultat fur Informatik, 2007.</mixed-citation></citation-alternatives></ref><ref id="cit10"><label>10</label><citation-alternatives><mixed-citation xml:lang="ru">Errica F., Podda M., Bacciu D., Micheli A. A fair comparison of graph neural networks for graph classification. Proc. of the 8th International Conference on Learning Representations, 2020.</mixed-citation><mixed-citation xml:lang="en">Errica F., Podda M., Bacciu D., Micheli A. A fair comparison of graph neural networks for graph classification. Proc. of the 8th International Conference on Learning Representations, 2020.</mixed-citation></citation-alternatives></ref><ref id="cit11"><label>11</label><citation-alternatives><mixed-citation xml:lang="ru">Gilmer J., Schoenholz S.S., Riley P.F., Vinyals O., Dahl G.E. Neural message passing for quantum chemistry. Proc. of the 34th International Conference on Machine Learning, 2017, vol. 70, pp. 1263–1272.</mixed-citation><mixed-citation xml:lang="en">Gilmer J., Schoenholz S.S., Riley P.F., Vinyals O., Dahl G.E. Neural message passing for quantum chemistry. Proc. of the 34th International Conference on Machine Learning, 2017, vol. 70, pp. 1263–1272.</mixed-citation></citation-alternatives></ref><ref id="cit12"><label>12</label><citation-alternatives><mixed-citation xml:lang="ru">Hamilton W.L., Ying R., Leskovec J. Inductive representation learning on large graphs. Proc. of the 31st International Conference on Neural Information Processing Systems, 2017, pp. 1025–1035.</mixed-citation><mixed-citation xml:lang="en">Hamilton W.L., Ying R., Leskovec J. Inductive representation learning on large graphs. Proc. of the 31st International Conference on Neural Information Processing Systems, 2017, pp. 1025–1035.</mixed-citation></citation-alternatives></ref><ref id="cit13"><label>13</label><citation-alternatives><mixed-citation xml:lang="ru">Xu D., Ruan Ch., Körpeoglu E., Kumar S., Achan K. Inductive representation learning on temporal graphs. Proc. of the 8th International Conference on Learning Representations, 2020.</mixed-citation><mixed-citation xml:lang="en">Xu D., Ruan Ch., Körpeoglu E., Kumar S., Achan K. Inductive representation learning on temporal graphs. Proc. of the 8th International Conference on Learning Representations, 2020.</mixed-citation></citation-alternatives></ref><ref id="cit14"><label>14</label><citation-alternatives><mixed-citation xml:lang="ru">Velickovic P., Cucurull G., Casanova A., Romero A., Liò P., Bengio Y. Graph attention networks. Proc. of the 6th International Conference on Learning Representations, 2018.</mixed-citation><mixed-citation xml:lang="en">Velickovic P., Cucurull G., Casanova A., Romero A., Liò P., Bengio Y. Graph attention networks. Proc. of the 6th International Conference on Learning Representations, 2018.</mixed-citation></citation-alternatives></ref><ref id="cit15"><label>15</label><citation-alternatives><mixed-citation xml:lang="ru">Kipf T.N., Welling M. Semi-supervised classification with graph convolutional networks. Proc. of the 5th International Conference on Learning Representations, 2017.</mixed-citation><mixed-citation xml:lang="en">Kipf T.N., Welling M. Semi-supervised classification with graph convolutional networks. Proc. of the 5th International Conference on Learning Representations, 2017.</mixed-citation></citation-alternatives></ref><ref id="cit16"><label>16</label><citation-alternatives><mixed-citation xml:lang="ru">Xu K., Hu W., Leskovec J., Jegelka S. How powerful are graph neural networks? Proc. of the 7th International Conference on Learning Representations, 2019.</mixed-citation><mixed-citation xml:lang="en">Xu K., Hu W., Leskovec J., Jegelka S. How powerful are graph neural networks? Proc. of the 7th International Conference on Learning Representations, 2019.</mixed-citation></citation-alternatives></ref><ref id="cit17"><label>17</label><citation-alternatives><mixed-citation xml:lang="ru">Leman A.A., Weisfeiler B. A reduction of a graph to a canonical form and an algebra arising during this reduction. Nauchno-Technicheskaya Informatsia. Seriya 2, 1968, no. 9, pp. 12–16.</mixed-citation><mixed-citation xml:lang="en">Leman A.A., Weisfeiler B. A reduction of a graph to a canonical form and an algebra arising during this reduction. Nauchno-Technicheskaya Informatsia. Seriya 2, 1968, no. 9, pp. 12–16.</mixed-citation></citation-alternatives></ref><ref id="cit18"><label>18</label><citation-alternatives><mixed-citation xml:lang="ru">Morris C., Ritzert M., Fey M., Hamilton W.L., Lenssen J.E., Rattan G., Grohe M. Weisfeiler and Leman go neural: Higher-order graph neural networks. Proceedings of the AAAI Conference on Artificial Intelligence, 2019, vol. 33, no. 1. pp. 4602–4609. https://doi.org/10.1609/aaai.v33i01.33014602</mixed-citation><mixed-citation xml:lang="en">Morris C., Ritzert M., Fey M., Hamilton W.L., Lenssen J.E., Rattan G., Grohe M. Weisfeiler and Leman go neural: Higher-order graph neural networks. Proceedings of the AAAI Conference on Artificial Intelligence, 2019, vol. 33, no. 1. pp. 4602–4609. https://doi.org/10.1609/aaai.v33i01.33014602</mixed-citation></citation-alternatives></ref><ref id="cit19"><label>19</label><citation-alternatives><mixed-citation xml:lang="ru">Maron H., Ben-Hamu H., Serviansky H., Lipman Y. Provably powerful graph networks. Proc. of the 33rd International Conference on Neural Information Processing Systems, 2019, pp. 2156–2167.</mixed-citation><mixed-citation xml:lang="en">Maron H., Ben-Hamu H., Serviansky H., Lipman Y. Provably powerful graph networks. Proc. of the 33rd International Conference on Neural Information Processing Systems, 2019, pp. 2156–2167.</mixed-citation></citation-alternatives></ref><ref id="cit20"><label>20</label><citation-alternatives><mixed-citation xml:lang="ru">Morris C., Kriege N.M., Bause F., Kersting K., Mutzel P., Neumann M. Tudataset: A collection of benchmark datasets for learning with graphs. arXiv, 2020. arXiv:2007.08663. https://doi.org/10.48550/arXiv.2007.08663</mixed-citation><mixed-citation xml:lang="en">Morris C., Kriege N.M., Bause F., Kersting K., Mutzel P., Neumann M. Tudataset: A collection of benchmark datasets for learning with graphs. arXiv, 2020. arXiv:2007.08663. https://doi.org/10.48550/arXiv.2007.08663</mixed-citation></citation-alternatives></ref><ref id="cit21"><label>21</label><citation-alternatives><mixed-citation xml:lang="ru">Järvelin K., Kekäläinen J. Cumulated gain-based evaluation of IR techniques. ACM Transactions on Information Systems, 2002, vol. 20, no. 4, pp. 422–446. https://doi.org/10.1145/582415.582418</mixed-citation><mixed-citation xml:lang="en">Järvelin K., Kekäläinen J. Cumulated gain-based evaluation of IR techniques. ACM Transactions on Information Systems, 2002, vol. 20, no. 4, pp. 422–446. https://doi.org/10.1145/582415.582418</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>
