Modeling, Reproduction, and Mapping of Geofields with and Without Measurement Noise. Part 4. Spline, Geostatistical, and Fuzzy Regression Analysis Techniques
- Authors: Pashayev A.M.1, Sadykhov R.A.1, Habibullayev S.B.1
-
Affiliations:
- National Academy of Aviation of Azerbaijan
- Issue: Vol 60, No 5 (2017)
- Pages: 432-442
- Section: Article
- URL: https://journals.rcsi.science/0543-1972/article/view/246157
- DOI: https://doi.org/10.1007/s11018-017-1214-3
- ID: 246157
Cite item
Abstract
Hybrid three-stage algorithms for identification of geofields that can yield high-quality indicators are examined. Integrable estimation techniques from the hybrid noise-resistant algorithm database are used for effective solution of geofield recovery problems with substantial measurement noise which are difficult to formulate and have multiple extremes. Fuzzy regression models are shown to lead to a multi-extremum non-convex optimization problem. An example is given of a single-extremum problem for finding a fuzzy linear regression model.
About the authors
A. M. Pashayev
National Academy of Aviation of Azerbaijan
Email: izmt@vniims.ru
Azerbaijan, Baku
R. A. Sadykhov
National Academy of Aviation of Azerbaijan
Email: izmt@vniims.ru
Azerbaijan, Baku
S. B. Habibullayev
National Academy of Aviation of Azerbaijan
Email: izmt@vniims.ru
Azerbaijan, Baku