Terminated in academic year 2022/2023

Methods of Experimental Data Processing

Type of study	Follow-up Master
Language of instruction	English
Code	548-0113/02
Abbreviation	MEZEK
Course title	Methods of Experimental Data Processing
Credits	5
Coordinating department	Department of Geoinformatics
Course coordinator	Ing. Lucie Orlíková, Ph.D.

Subject syllabus

1)Introduction to artificial neural networks
2) Architecture of artificial neural networks. Perceptrons and basic learning algorithms
3) Backpropagation learning
4) Competitive Learning and Kohonen Nets
5) CounterPropagation method
6) Hopfield Nets and Boltzmann Machines
7) Optimization Techniques, overfitting, cross validation
8) Support vector classification
9) Support vector machine - kernel methods
10) Artificial neural networks in geoinformatics

Literature

HAYKIN, Simon S. Neural networks and learning machines. 3rd ed. Upper Saddle River: Pearson, 2009. 934 s. ISBN 9780131293762.
KOHONEN, Teuvo. Self-Organizing Maps. Berlin: Springer-Verlag, 1995. 392 s. Springer Series in Information Sciences 30. . info
Kanevski M., Pozdnoukhov A., Timonin V. Machine Learning for Spatial Environmental Data. EPFL Press, 2009.
Bishop C. Pattern recognition and machine learning. Springer, 2006.

Advised literature

Hastie T., Tibshirani R., Friedman J. The Elements of Statistical Learning. 2d edition. Springer, 2009.
Kanevski M. (Editor). Advanced Mapping of Environmental Data. Geostatistics, Machine Learning, and Bayesian Maximum Entropy. iSTE/Wiley, 2008.

back to search page