Spatial dependence between training and test sets: another pitfall of classification accuracy assessment in remote sensing

DOI10.1007/S10994-021-05972-1MaRDI QIDQ2102335zbMATH OpenOpenAlexFDO

Authors N. Karasiak, J.-F. Dejoux, C. Monteil, D. Sheeren

Publication date 28 November 2022

Published in Machine Learning (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1007/s10994-021-05972-1

zbMATH Keywords

cross-validationremote sensingspatial autocorrelationoverfittingaccuracy assessment

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05)

Recommendations

• The influence of training sampling size on the expected error rate in spatial classification
• Error rates in spatial classification of Gaussian data with random labeling
• Comparison of spatial classification rules with different conditional distributions of class label

Cites work

• A test for spatial autocorrelation in seemingly unrelated regressions
• Random forests
• Scikit-learn: machine learning in Python

Cited in

(1)

• Museo ToolBox

Describes a project that uses

Uses Software

• Scikit

This page was built for publication: Spatial dependence between training and test sets: another pitfall of classification accuracy assessment in remote sensing

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2102335)