Autocorrelation in spatial regression with Machine Learning methods (with a focus on Random Forest)

Short talk for the workshop Hands on Machine Learning in Ecology organized by Marta M. Rufino (MARE, Faculty of Sciences, University of Lisbon) and Tiago A. Marques (University of St. Andrews).

In this talk I focus on the nature and meaning of spatial autocorrelation, and describe how it can be incorporated into spatial regression models fitted with random forest models with the R package spatialRF.

Link to slideshow