Analysis of the Inerka polygon metageosystems by means of Ensembles of machine learning models

DOI: 10.35595/2414-9179-2022-1-28-613-628

View or download the article (Rus)

About the Authors

Anatoliy A. Yamashkin

National Research Mordovia State University,
ul. Bolshevistskaya 68, 430005, Saransk, Russia;
E-mail: yamashkinsa@mail.ru

Stanislav A. Yamashkin

National Research Mordovia State University,
ul. Bolshevistskaya 68, 430005, Saransk, Russia;
E-mail: yamashkin56@mail.ru

Abstract

The article describes a geoinformation algorithm for interpreting Earth remote sensing data based on the Ensemble Learning methodology. The proposed solution can be used to assess the stability of geosystems and predict natural (including exogeodynamic) processes. The difference of the created approach is determined by a fundamentally new organization scheme of the metaclassifier as a decision-making unit, as well as the use of a geosystem approach to preparing data for automated analysis using deep neural network models. The article shows that the use of ensembles, built according to the proposed method, makes it possible to carry out an operational automated analysis of spatial data for solving the problem of thematic mapping of metageosystems and natural processes. At the same time, combining models into an ensemble based on the proposed architecture of the metaclassifier makes it possible to increase the stability of the analyzing system: the accuracy of decisions made by the ensemble tends to tend to the accuracy of the most efficient monoclassifier of the system. The integration of individual classifiers into ensembles makes it possible to approach the solution of the scientific problem of finding classifier hyperparameters through the combined use of models of the same type with different configurations. The formation of a metaclassifier according to the proposed algorithm is an opportunity to add an element of predictability and control to the use of neural network models, which are traditionally a “black box”. Mapping of the geosystems of the Inerka test site shows their weak resistance to recreational development. The main limiting factors are the composition of Quaternary deposits, the nature of the relief, the mechanical composition of soils, soil moisture, the thickness of the humus horizon of the soil, the genesis and composition of vegetation.

Keywords

ensembles, machine learning, metageosystems, cartography, sustainable development, GIS

References

  1. Boucher M., Perreault L., Anctil F. Tools for the assessment of hydrological ensemble forecasts obtained by neural networks. Journal of Hydroinformatics, 2009, V. 11, No. 3–4. P. 297–307. DOI: 10.2166/hydro.2009.037.
  2. Dong X., Yu Z., Cao W., Shi Y., Ma Q. A survey on ensemble learning. Frontiers of Computer Science, 2020, V. 14, No. 2. P. 241–258. DOI: 10.1007/s11704-019-8208-z.
  3. Gkonos C., Iosifescu Enescu I., Hurni L. Spinning the wheel of design: evaluating geoportal Graphical User Interface adaptations in terms of human-centred design. International Journal of Cartography, 2019, V. 5, No. 1, P. 23–43. DOI: 10.1080/23729333.2018.1468726.
  4. Han Q., Zhao N., Xu J. Recognition and location of steel structure surface corrosion based on unmanned aerial vehicle images. Journal of Civil Structural Health Monitoring, 2021, V. 11, No. 5, P. 1375–1392. DOI: 10.1007/s13349-021-00515-7.
  5. Heaton J., Datta A., Finley A.O. A case study competition among methods for analyzing large spatial data. Journal of Agricultural, Biological and Environmental Statistics, 2019, V. 24, No. 3, P. 398–425. DOI: 10.1007/s13253-018-00348-w.
  6. Kim S.E., Seo I.W. Artificial neural network ensemble modeling with exploratory factor analysis for streamflow forecasting. Journal of Hydroinformatics, 2015, V. 17, No. 4. P. 614–639. DOI: 10.2166/hydro.2015.033.
  7. Kuznetsov A.V., Myasnikov V.V. Comparison of algorithms for controlled element-by-element classification of hyperspectral images. Computer Optics, 2014, V. 38, No. 3. P. 494–502. (in Russian).
  8. LeCun Y., Bengio Y., Hinton G. Deep learning, Nature, 2015, V. 521, No. 7553, P. 436–444. DOI: doi.org/10.1038/nature14539.
  9. Lee J., Kang M. Geospatial Big Data: Challenges and Opportunities, Big Data Research, 2017, V. 2, No. 2, P. 74–81. DOI: 10.1016/j.bdr.2015.01.003.
  10. Nikolaev V.A. Classification and small-scale landscape mapping. Moscow: Moscow State University Press, 1978. 62 p. (in Russian).
  11. Sergeev V.V., Yuzkiv R.R. Parametric model of the autocorrelation function of space hyperspectral images. Computer Optics, 2016, V. 40, No. 3, P. 416–421. (in Russian).
  12. Tikunov V.S., Kotova T.V., Belousov S.K. Environmental conditions: definition, indicators, mapping. InterCarto. InterGIS, 2021, V. 27. Part 1. P. 165–194 (in Russian). DOI: 10.35595/2414-9179-2021-1-27-165-194.
  13. Yamashkin S.A., Yamashkin A.A., Zanozin V.V., Radovanovic M.M., Barmin A.N. Improving the efficiency of deep learning methods in remote sensing data analysis: geosystem approach. IEEE Access, 2020, V. 8. P. 179516–179529. DOI: 10.1109/ACCESS.2020.3028030.

For citation: Yamashkin A.A., Yamashkin S.A. Analysis of the Inerka polygon metageosystems by means of Ensembles of machine learning models. InterCarto. InterGIS. GI support of sustainable development of territories: Proceedings of the International conference. Moscow: MSU, Faculty of Geography, 2022. V. 28. Part 1. P. 613–628. DOI: 10.35595/2414-9179-2022-1-28-613-628 (in Russian)