<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE root>
<article 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" xmlns:ali="http://www.niso.org/schemas/ali/1.0/" article-type="research-article" dtd-version="1.2" xml:lang="en"><front><journal-meta><journal-id journal-id-type="publisher-id">Computational nanotechnology</journal-id><journal-title-group><journal-title xml:lang="en">Computational nanotechnology</journal-title><trans-title-group xml:lang="kk"><trans-title>Computational nanotechnology</trans-title></trans-title-group><trans-title-group xml:lang="pt"><trans-title>Computational nanotechnology</trans-title></trans-title-group><trans-title-group xml:lang="ru"><trans-title>Computational nanotechnology</trans-title></trans-title-group><trans-title-group xml:lang="zh"><trans-title>Computational nanotechnology</trans-title></trans-title-group></journal-title-group><issn publication-format="print">2313-223X</issn><issn publication-format="electronic">2587-9693</issn><publisher><publisher-name xml:lang="en">YUR-VAK</publisher-name></publisher></journal-meta><article-meta><article-id pub-id-type="publisher-id">688950</article-id><article-id pub-id-type="doi">10.33693/2313-223X-2025-12-2-11-18</article-id><article-id pub-id-type="edn">QFHIYD</article-id><article-categories><subj-group subj-group-type="toc-heading" xml:lang="en"><subject>ARTIFICIAL INTELLIGENCE AND MACHINE LEARNING</subject></subj-group><subj-group subj-group-type="toc-heading" xml:lang="ru"><subject>ИСКУССТВЕННЫЙ ИНТЕЛЛЕКТ И МАШИННОЕ ОБУЧЕНИЕ</subject></subj-group><subj-group subj-group-type="article-type"><subject>Research Article</subject></subj-group></article-categories><title-group><article-title xml:lang="en">Improving time series forecasting by applying the sliding window approach</article-title><trans-title-group xml:lang="ru"><trans-title>Повышение точности моделей прогнозирования временных рядов с помощью метода скользящего окна</trans-title></trans-title-group></title-group><contrib-group><contrib contrib-type="author"><contrib-id contrib-id-type="orcid">https://orcid.org/0000-0002-0389-9758</contrib-id><contrib-id contrib-id-type="scopus">58106191000</contrib-id><contrib-id contrib-id-type="researcherid">rid66654</contrib-id><name-alternatives><name xml:lang="en"><surname>Ndungi</surname><given-names>Rebeccah</given-names></name><name xml:lang="ru"><surname>Ндунги</surname><given-names>Ребекка</given-names></name></name-alternatives><address><country country="RU">Russian Federation</country></address><bio xml:lang="en"><p>postgraduate student, Faculty of Mathematics and Computer Science</p></bio><bio xml:lang="ru"><p>аспирант, факультет математики и компьютерных наук</p></bio><email>Rebeccahndungi94@gmail.com</email><xref ref-type="aff" rid="aff1"/></contrib><contrib contrib-type="author"><contrib-id contrib-id-type="orcid">https://orcid.org/0000-0002-7305-1429</contrib-id><contrib-id contrib-id-type="scopus">56149559700</contrib-id><contrib-id contrib-id-type="researcherid">G-8844-2015</contrib-id><contrib-id contrib-id-type="spin">7473-1900</contrib-id><name-alternatives><name xml:lang="en"><surname>Blekanov</surname><given-names>Ivan S.</given-names></name><name xml:lang="ru"><surname>Блеканов</surname><given-names>Иван Станиславович</given-names></name></name-alternatives><address><country country="RU">Russian Federation</country></address><bio xml:lang="en"><p>Cand. Sci. (Eng.), Associate Professor; Head, Department of Programming Technology</p></bio><bio xml:lang="ru"><p>кандидат технических наук, доцент; заведующий, кафедра технологий программирования</p></bio><email>I.blekanov@spbu.ru</email><xref ref-type="aff" rid="aff1"/></contrib></contrib-group><aff-alternatives id="aff1"><aff><institution xml:lang="en">St. Petersburg State University</institution></aff><aff><institution xml:lang="ru">Санкт-Петербургский государственный университет</institution></aff></aff-alternatives><pub-date date-type="pub" iso-8601-date="2025-08-19" publication-format="electronic"><day>19</day><month>08</month><year>2025</year></pub-date><volume>12</volume><issue>2</issue><issue-title xml:lang="en"/><issue-title xml:lang="ru"/><fpage>11</fpage><lpage>18</lpage><history><date date-type="received" iso-8601-date="2025-08-11"><day>11</day><month>08</month><year>2025</year></date></history><permissions><copyright-statement xml:lang="en">Copyright ©; 2025, Yur-VAK</copyright-statement><copyright-statement xml:lang="ru">Copyright ©; 2025, Юр-ВАК</copyright-statement><copyright-year>2025</copyright-year><copyright-holder xml:lang="en">Yur-VAK</copyright-holder><copyright-holder xml:lang="ru">Юр-ВАК</copyright-holder><ali:free_to_read xmlns:ali="http://www.niso.org/schemas/ali/1.0/" start_date="2026-08-19"/><license><ali:license_ref xmlns:ali="http://www.niso.org/schemas/ali/1.0/">https://www.urvak.ru/contacts/</ali:license_ref></license></permissions><self-uri xlink:href="https://journals.eco-vector.com/2313-223X/article/view/688950">https://journals.eco-vector.com/2313-223X/article/view/688950</self-uri><abstract xml:lang="en"><p>Our<bold> </bold>primary research involves forecasting the IT job market, where we study trends, residuals, and seasonalities. In this study, we<bold> </bold>focus on the impact of the sliding window technique on forecasting models. The sliding window approach in the machine learning process is aimed at enhancing the accuracy of the forecasting models. It involves partitioning the continuous time series into subsets of consecutive and overlapping periods, which enables the models to track temporal characteristics effectively. The experiment is carried out on various algorithms and integrated with the sliding window. The technique allows flexibility for models to adapt to changes in the data dynamics, which greatly reduces the errors in forecasting. The study shows that sliding window methods are quite useful for building dependable and adaptive forecasting models. LSTM, ARIMA, SARIMA, and Holt’s Model were used in this experiment with a dataset of 1<sup> </sup>048<sup> </sup>576 job rows with job-related information. Metrics such as MSE, RMSE, and MAE were used to test the models. LSTM was found to be the most efficient because of its capability to learn complicated patterns and long-term dependencies, and showed model improvement of 0.248 on MAE, 2.649 on MSE, and 0.162 on RMSE when the sliding window was applied.</p></abstract><trans-abstract xml:lang="ru"><p>Данное исследование посвящено методам прогнозирования рынка труда ИТ-вакансий, в рамках которого изучаются вопросы выявления трендов, анализа остатков и сезонности временных рядов. В данном работе авторы основное внимание уделяют влиянию метода скользящего окна на модели прогнозирования временного ряда. Использование такого подхода предобработки данных направлен на повышение точности моделей машинного обучения для прогнозирования временных рядов. Рассматриваемый метод основывается на разбиении непрерывного временного ряда на множество последовательных и пересекающихся периодов с фиксированным размером, что позволяет моделям прогнозирования эффективно отслеживать временные характеристики. В работа был проведен эксперимент по оценке влияния использования метода скользящего окна в сочетании с различными моделями прогнозирования на качество прогноза ИТ-вакансий на наборе данных, содержащем 1<sup> </sup>048<sup> </sup>576 строк с информацией о вакансиях. В эксперименте в качестве моделей прогнозирования использовались LSTM, ARIMA, SARIMA и модель Холта. Для оценки качества моделей применялись метрики MSE, RMSE и MAE. Такое сочетание техники предобработки и модели обеспечивает устойчивость качества прогноза к чувствительности данных и адаптированность к резким изменениям в данных временного ряда, что значительно снижает ошибки прогнозирования по качественным показателям. Эксперимент показал, что модель LSTM оказалась наиболее эффективной благодаря способности более глубоко изучать сложные закономерности и выявлять долгосрочные зависимости, продемонстрировав прирост качества базовой модели с использованием метода скользящего окна по метрике MAE на 0,248 условные единицы, по MSE – на 2,649, по RMSE – на 0,162, по сравнению с базовой моделью (без скользящего окна). Таким образом, авторы работы показывают, что методы скользящего окна весьма полезны для построения устойчивых и адаптивных моделей прогнозирования.</p></trans-abstract><kwd-group xml:lang="en"><kwd>sliding window</kwd><kwd>time series forecasting</kwd><kwd>machine learning</kwd><kwd>IT job market</kwd><kwd>trends and seasonality</kwd><kwd>model accuracy</kwd></kwd-group><kwd-group xml:lang="ru"><kwd>скользящее окно</kwd><kwd>прогнозирование временных рядов</kwd><kwd>машинное обучение</kwd><kwd>рынок ИТ-вакансий</kwd><kwd>тренды и сезонность</kwd><kwd>точность модели</kwd></kwd-group><funding-group/></article-meta></front><body></body><back><ref-list><ref id="B1"><label>1.</label><mixed-citation>Alsharef A., Aggarwal K., Garg S. et al. Review of ML and AutoML solutions to forecast time-series data. Archives of Computational Methods in Engineering. 2022. Vol. 29. No. 7. Pp. 5297–5311. DOI: 10.1007/S11831-022-09765-0/METRICS.</mixed-citation></ref><ref id="B2"><label>2.</label><mixed-citation>Fatima S.S.W., Rahimi A. A Review of time-series forecasting algorithms for industrial manufacturing systems. Machines. 2024. Vol. 12. No. 6. P. 380. DOI: 10.3390/MACHINES12060380.</mixed-citation></ref><ref id="B3"><label>3.</label><mixed-citation>Itzhak N., Jaroszewicz S., Moskovitch R. Temporal ensemble of multiple patterns’ instances for continuous prediction of events. Mach. Learn. 2025. Vol. 114. No. 5. Pp. 1–42. DOI: 10.1007/S10994-025-06756-7/FIGURES/16.</mixed-citation></ref><ref id="B4"><label>4.</label><mixed-citation>Christensen H.B., Hail L., Leuz C. Mandatory CSR and sustainability reporting: Economic analysis and literature review. Review of Accounting Studies. 2021. Vol. 26. No. 3. Pp. 1176–1248. DOI: 10.1007/S11142-021-09609-5.</mixed-citation></ref><ref id="B5"><label>5.</label><mixed-citation>Kontopoulou V.I., Panagopoulos A.D., Kakkos I., Matsopoulos G.K. A review of ARIMA vs. machine learning approaches for time series forecasting in data driven networks. Future Internet. 2023. Vol. 15. No. 8. P. 255. DOI: 10.3390/FI15080255.</mixed-citation></ref><ref id="B6"><label>6.</label><mixed-citation>Forecasting labor and skill demand by sector and occupation. In 2 vols. Vol. 1: Case studies and guidance. URL: https://documents.worldbank.org/en/publication/documents-reports/documentdetail/251591531754581450/case-studies-and-guidance (data of accesses: 19.04.2025).</mixed-citation></ref><ref id="B7"><label>7.</label><mixed-citation>Yurtsever M. Unemployment rate forecasting: LSTM-GRU hybrid approach. J. Labour. Mark. Res. 2023. Vol. 57. No. 1. Pp. 1–9. DOI: 10.1186/S12651-023-00345-8/FIGURES/5.</mixed-citation></ref><ref id="B8"><label>8.</label><mixed-citation>Vorobev A.V., Kudinov V.A. The importance of forecasting in industrial enterprise management using machine learning. Scientific and Technical Information Processing. 2022. Vol. 49. No. 5. Pp. 393–398. DOI: 10.3103/S0147688222050173.</mixed-citation></ref><ref id="B9"><label>9.</label><mixed-citation>Hewamalage H., Ackermann K., Bergmeir C. Forecast evaluation for data scientists: Common pitfalls and best practices. Data Min Knowl Discov. 2023. Vol. 37. No. 2. Pp. 788–832. DOI: 10.1007/S10618-022-00894-5/FIGURES/14.</mixed-citation></ref><ref id="B10"><label>10.</label><mixed-citation>Lukats D., Zielinski O., Hahn A., Stahl F. A benchmark and survey of fully unsupervised concept drift detectors on real-world data streams. Int. J. Data Sci. Anal. 2024. Vol. 19. No. 1. Pp. 1–31. DOI: 10.1007/S41060-024-00620-Y/FIGURES/13.</mixed-citation></ref><ref id="B11"><label>11.</label><mixed-citation>Zhang J., Zeng Y., Starly B. Recurrent neural networks with long-term temporal dependencies in machine tool wear diagnosis and prognosis. SN Appl. Sci. 2021. Vol. 3. No. 4. Pp. 1–13.. DOI: 10.1007/S42452-021-04427-5/FIGURES/7.</mixed-citation></ref><ref id="B12"><label>12.</label><mixed-citation>Hyndman R.J., Khandakar Y. Automatic time series forecasting: The forecast package for R. J. Stat. Softw. 2008. Vol. 27. No. 3. Pp. 1–22. DOI: 10.18637/JSS.V027.I03.</mixed-citation></ref><ref id="B13"><label>13.</label><mixed-citation>Forecasting: Principles and practice. 3rd ed. Accessed: Nov. 14, 2024. URL: https://otexts.com/fpp3/ (data of accesses: 14.11.2024).</mixed-citation></ref><ref id="B14"><label>14.</label><mixed-citation>Fokianos K., Fried R. Interventions in log-linear Poisson autoregression. Statistical Modelling. 2012. Vol. 12. No. 4. Pp. 299–322. DOI: 10.1177/1471082X1201200401.</mixed-citation></ref><ref id="B15"><label>15.</label><mixed-citation>Zhang G., Patuwo B.E., Hu M.Y. Forecasting with artificial neural networks: The state of the art. Int. J. Forecast. 1998. Vol. 14. No. 1. Pp. 35–62. DOI: 10.1016/S0169-2070(97)00044-7.</mixed-citation></ref><ref id="B16"><label>16.</label><mixed-citation>Vafaeipour M., Rahbari O., Rosen M.A. et al. Application of sliding window technique for prediction of wind velocity time series. International Journal of Energy and Environmental Engineering. 2014. Vol. 5. No. 2–3. Pp. 1–7. DOI: 10.1007/S40095-014-0105-5/FIGURES/7.</mixed-citation></ref><ref id="B17"><label>17.</label><mixed-citation>Chen Y., Fan X., Huang R. et al. Choose an item. Artificial intelligence/machine learning technology in power system applications. 2024. URL: www.osti.gov (data of accesses: 19.04.2025).</mixed-citation></ref><ref id="B18"><label>18.</label><mixed-citation>Bressane A., Garcia A.J. da S., de Castro M.V. et al. Fuzzy machine learning applications in environmental engineering: Does the ability to deal with uncertainty matter? Sustainability. 2024. Vol. 16. No. 11. P. 4525. DOI: 10.3390/SU16114525.</mixed-citation></ref><ref id="B19"><label>19.</label><mixed-citation>Wu J.M.T., Li Z., Herencsar N. et al. A graph-based CNN-LSTM stock price prediction algorithm with leading indicators. Multimed. Syst. 2023. Vol. 29. No. 3. Pp. 1751–1770. DOI: 10.1007/S00530-021-00758-W/TABLES/15.</mixed-citation></ref><ref id="B20"><label>20.</label><mixed-citation>Tao Z., Xu Q., Liu X., Liu J. An integrated approach implementing sliding window and DTW distance for time series forecasting tasks. Applied Intelligence. 2023. Vol. 53. No. 17. Pp. 20614–20625. DOI: 10.1007/S10489-023-04590-9/METRICS.</mixed-citation></ref></ref-list></back></article>
