Correcting the effect of sampling bias in species distribution modeling - A new method in the case of a low number of presence data.
Yi MouaEmmanuel RouxFrédérique SeylerSébastien BriolantPublished in: Ecol. Informatics (2020)
Keyphrases
- synthetic data
- random samples
- data sets
- noisy data
- input data
- training data
- small number
- correlation analysis
- spectral clustering
- prior knowledge
- prior information
- data distribution
- computational complexity
- data analysis
- modeling method
- original data
- test data
- missing data
- probability distribution function
- data sources
- preprocessing
- high dimensional data
- probability distribution
- k means
- pairwise
- density function
- random sampling
- information loss
- highly correlated
- data structure
- random sample
- database
- objective function