Understanding the role of training sample size in the uncertainty of high-resolution LULC mapping using random forest.
Kwanele PhinziNjoya Silas NgetarQuoc Bao PhamGashaw Gismu ChakiluSzilárd SzabóPublished in: Earth Sci. Informatics (2023)
Keyphrases
- sample size
- random forest
- high resolution
- random forests
- land cover
- decision trees
- model selection
- remote sensing
- upper bound
- feature set
- image processing
- random sampling
- multi label
- ensemble methods
- worst case
- multispectral
- learning algorithm
- progressive sampling
- prediction accuracy
- satellite images
- training data
- text classification
- supervised learning
- training set
- pattern recognition
- base classifiers
- optimal solution
- similarity measure