<scp>itsdm</scp> : Isolation forest‐based presence‐only species distribution modelling and explanation in <scp>r</scp>

Multiple statistical algorithms have been used for species distribution modelling (SDM). Due to shortcomings in occurrence datasets, presence-only methods (such as MaxEnt) become increasingly widely used. However, sampling bias remains a challenging issue, particularly density-based approaches. The Isolation Forest (iForest) algorithm is method less sensitive patterns and over-fitting because it fits the model by describing unsuitable instead of suitable conditions. Here, we present itsdm package with iForest, which provides workflow wrapper iForest family convenient tools diagnostic post-modelling analysis. allows users fit evaluate an SDM using data. It also helps understand relationships between living environment Shapley values, suggested technique explainable artificial intelligence (xAI). Additionally, can make spatial response maps that indicate how respond environmental variables across space detect areas potentially affected changing environment. We demonstrated usage compared other mainstream SDMs virtual species. results enlightened advantageous when actual range unclear.