Multi-label Classification for Species Distribution Modeling.
Published in The International Conference on Machine Learning (ICML) Workshop, 2011
Species distribution models play a key role in creating reserves for species conservation, predicting the effects of ecological change, and testing ecological theory. Although many methods have been developed for models of individual species, ecologists are recognizing the advantages of predicting multiple species simultaneously. This problem of multiple species prediction can be addressed by machine learning algorithms from the area of multi-label classification. However, to date, multi-label classification has been applied primarily to problems in text and image annotation. The goal of this paper is to introduce species distribution modeling as a new domain for multi-label classification, present preliminary results illustrating the advantages of multi-label algorithms, and discuss new research directions presented by this domain.