Modelling ecological niches with support vector machines

By John M. Drake, C. Randin, Antoine Guisan. 2006

340 citations recorded by the source corpus.

Subfields: Ecological Modeling, Ecology, Evolution, Behavior and Systematics, Nature and Landscape Conservation

View Paper: https://www.semanticscholar.org/paper/54dd013c04a3f588cb366d683a3defd947db3e07

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Ecological Modeling

In Ecology, Evolution, Behavior and Systematics

In Nature and Landscape Conservation

Citations in the corpus

Application of machine learning methods to spatial interpolation of environmental variables (2011, methodology): “…methods, like random forest (RF) and support vector machine (SVM), have shown their predictive accuracy in data mining fields and other disciplines in terms of predictive errors (Cutler et al., 2007; Diaz-Uriarte and de Andres, 2006; Drake et al., 2006; Marmion et al., 2009; Shan et al., 2006).”
Species Distribution Modeling in the Tropics: Problems, Potentialities, and the Role of Biological Data for Effective Species Conservation (2009, background): “…al. (2002) France Methodological Non reported Invertebrates P-A Domínguez-Domínguez et al. (2006) Mexico Species conservation Partitioning Fishes P Drake et al. (2006) Switzerland Methodological Partitioning Plants P Elith & Leathwick (2007) Australia, Canada, New Zealand, Switzerland, South…”
On the selection of thresholds for predicting species occurrence with presence‐only data (2015, methodology): “1993), LIVES (Li and Hilbert 2008), and one-class SVM (Guo et al. 2005; Drake et al. 2006), while other techniques use both presence data and accompanying pseudo-absence data (often random points) that represent the background context for the model.”
Effects of the number of presences on reliability and stability of MARS species distribution models: the importance of regional niche variation and ecological heterogeneity (2010, methodology): “2007), ten using GARP (Stockwell & Peterson 2002), 15 using GARP and Maxent (Papeş & Gaubert 2007), 20 using logistic multiple regression (Stockwell & Peterson 2002), 40 using support vector machines (Drake et al. 2006), more than 30 using 12 different methods included GARP and Maxent (Wisz et al.”
Species Distribution Models: Ecological Explanation and Prediction Across Space and Time (2009, neutral): “…Frescino 2002), classification and regression trees and ensembles of trees (random forests: Prasad et al. 2006; boosted regression trees: Elith et al. 2008), genetic algorithms (Stockwell & Peters 1999), support vector machines (Drake et al. 2006), and maximum entropy models (Phillips et al. 2006).”
A framework for species distribution modelling with improved pseudo-absence generation (2015, methodology): “To this aim, we run one–class sup-175 port vector machines (OCSVM, Scholkopf and Smola, 2001) for each Oak176 group (see e.g. Drake et al., 2006; Bedia et al., 2011, for specific details on the177 use of support vector machines in SDM studies).”
Is Cassava the Answer to African Climate Change Adaptation? (2012, background): “2003), support vector machines (Drake et al. 2006), among others.”
Measuring the accuracy of species distribution models: a review (2009, methodology): “This measure has been used in SDM (e.g. Drake et al., 2006 ).”
Species distribution modelling of marine benthos: a North Sea case study (2011, background): “With the help of a kernel function, the data points (representing known presence and absence records in the ecospace) can be mapped in a higher dimensional space in which complicated patterns can be more simply represented (Drake et al. 2006).”
Machine learning and deep learning—A review for ecologists (2022, background): “…researchers can benefit from the wide range of different ML algorithms (Figure 4 ): SVMs or kNN can handle sparse datasets well (Como et al., 2017 ; Drake et al., 2006 ) and LASSO, Ridge, and elastic-net regressions are well suited for datasets with more features than observations (Zou & Hastie,…”