Russian Federation
UDC 614.8
This article examines the complexity of probabilistic wildfire forecasting using machine learning. A training set including meteorological, satellite, and historical data is analyzed, distributions are visualized, and dimensionality reduction is performed using nonlinear t-SNE and UMAP methods. It is demonstrated that the "fire" and "no fire" classes significantly overlap in the feature space, suggesting the need for complex, nonlinear machine learning methods.
wildfires, machine learning, t-SNE, UMAP, data visualization
1. Matiev R.T., Matveev. A.V., Tarancev A.A. Model` podderzhki prinyatiya reshenij pri reagirovanii na lesnye pozhary v gornoj mestnosti s ispol`zovaniem nechetkoj logiki // Inzhenerny`j vestnik Dona. 2025. № 12(132). EDN LCTAWP.
2. Medvedev D.V., Matveev A.V. Algoritmy intellektual`noj podderzhki prinyatiya upravlencheskix reshenij pri ugrozax lesny`x pozharov // Nauchno-analiticheskij zhurnal Vestnik Sankt-Peterburgskogo universiteta Gosudarstvennoj protivopozharnoj sluzhby` MChS Rossii. 2025. № 2. S. 35-48. DOI:https://doi.org/10.61260/2218-130X-2025-2-35-48. EDN OKGHLE.
3. Matveev A.V., Matiev R.T. Prinyatie reshenij pri pozharax v gornoj mestnosti: Sravnitel`nyj analiz metodov monitoringa // Nacional`naya bezopasnost` i strategicheskoe planirovanie. 2023. № 2(42). S. 76-90. DOI:https://doi.org/10.37468/2307-1400-2023-2-76-90. EDN QPRUWC.
4. P. Jain [et al.] A review of machine learning applications in wildfire science and management // Environmental Reviews. 2020. V. 28. No 4. P. 478-505. DOI:https://doi.org/10.1139/er-2020-0019
5. Wildfire Risk Prediction: A Survey of Recent Advances Using Deep Learning Techniques / Z Xu [et al.] // 2024. arXiv:2405.01607v4.
6. Medvedev D.V., Matveev A.V. Informacionnaya model` podderzhki prinyatiya reshenij po reagirovaniyu na landshaftnye pozhary // Sibirskij pozharno-spasatel`ny`j vestnik. 2025. № 1(36). S. 117-125. DOI:https://doi.org/10.34987/vestnik.sibpsa.2025.22.36.011. EDN PKSBTR.
7. Medvedev D.V., Matveev A.V., Smirnov A.S. Primenenie modeli logisticheskoj regressii pri prinyatii reshenij po opredeleniyu kolichestva privlekaemyx sil na likvidaciyu lesnyx pozharov // Pozharovzryvobezopasnost`. 2024. T. 33. № 4. S. 84-96. DOI:https://doi.org/10.22227/0869-7493.2024.33.04.84-96. EDN MJLVTY.
8. Medvedev D.V., Matveev A.V., Dmitrieva A.I. Primenenie nejro-nechetkoj sistemy ANFIS pri prognozirovanii ploshhadi lesny`x pozharov // Pozharnaya bezopasnost`: sovremennye vyzovy. Problemy i puti resheniya: sb. materialy` Vseros.j nauch.-prakt. konf., Sankt-Peterburg, 18 aprelya 2024 goda. Sankt-Peterburg: Sankt-Peterburgskij universitet GPS MChS Rossii, 2024. S. 43-48. EDN PFZVXO.
9. Svidetel`stvo o gosudarstvennoj registracii programmy` dlya E`VM № 2024682958 Rossijskaya Federaciya. Nejro-nechetkaya sistema ANFIS dlya prognozirovaniya pokazatelej lesny`x pozharov : № 2024681214 : zayavl. 13.09.2024 : opubl. 01.10.2024 / D.V. Medvedev, A.V. Matveev; zayavitel` Sankt-Peterburgskij universitet GPS MChS Rossii. EDN MMAXOR.
10. Van der Maaten L., Hinton G. Visualizing data using t-SNE // Journal of machine learning research. 2008. V. 9. No. 11.
11. McInnes L., Healy J., Melville J. Umap: Uniform manifold approximation and projection for dimension reduction // arXiv preprint arXiv: 1802.03426. 2018. DOIhttps://doi.org/10.48550/arXiv.1802.03426
12. Scikit-learn: T SNE. URL: https://scikit-learn.org/stable/modules/manifold.html#t-distributed-stochastic-neighbor-embedding-t-sne/ (data obrashcheniya: 06.02.2026).
13. Umap-learn documentation. URL: https://umap-learn.readthedocs.io/ (data obrashcheniya: 06.02.2026).
14. CatBoost: unbiased boosting with categorical features / L. Prokhorenkova [et al.] // Advances in Neural Information Processing Systems. 2018. Vol. 31.
15. Hochreiter S., Schmidhuber J. Long Short Term Memory // Neural Computation. 1997. Vol. 9. № 8. P. 1735–1780. DOI:https://doi.org/10.1162/neco.1997.9.8.1735
