Artificial Neural Networks (ANN) as a predictive modelling tool.

(Below is a short introduction of ANN I wrote for our project report. I knew pretty much nothing about ANN last week. This short essay is based on my readings on about 10 papers.)

Decision makers in charge of ecosystem management constantly face the problem of analysing complex relationships that we currently lack of a prior understanding of. Under such circumstances ANNs have proved to be more powerful than traditional statistical approaches (Lek, Delacoste et al. 1996; Paruelo and Tomasel 1997).

Based on function of the human brain ANN is a non-linear mapping approach that has great capacity in predictive modelling (Lek and Guegan 1999). During the last decade or so ANNs have been widely applied in prediction of functional characteristics of ecosystems (Paruelo and Tomasel 1997); of global riverine fish diversity (Guegan, Lek et al. 1998), of environmental impact due to wildlife damage (Spitz and Lek 1999), of insect species richness (Park, Cereghino et al. 2003) and of freshwater community composition (Olden, Joy et al. 2006).

Only recently has ANN been used in predicting biological invasion (Gevrey, Worner et al. 2006; Gevrey and Worner 2006; Worner and Gevrey 2006). In such studies pest species assemblages are regarded as non-random species groupings that contain hidden predictive information. Such information can assist in identifying species that have the potential to pose an invasive threat in regions where they are not normally found. The underlying assumption is that geographical areas with similar pest assemblages share similar biotic and abiotic conditions that allow particular pest species to invade the area (Worner and Gevrey 2006).

References

Gevrey, M., S. Worner, et al. (2006). "Estimating risk of events using SOM models: A case study on invasive species establishment." Ecological Modelling 197(3-4): 361-372.

Gevrey, M. and S. P. Worner (2006). "Prediction of global distribution of insect pest species in relation to climate by using an ecological informatics method." Journal of Economic Entomology 99(3): 979-986.

Guegan, J. F., S. Lek, et al. (1998). "Energy availability and habitat heterogeneity predict global riverine fish diversity." Nature 391(6665): 382-384.

Lek, S., M. Delacoste, et al. (1996). "Application of neural networks to modelling nonlinear relationships in ecology." Ecological Modelling 90(1): 39-52.

Lek, S. and J. F. Guegan (1999). "Artificial neural networks as a tool in ecological modelling, an introduction." Ecological Modelling 120(2-3): 65-73.

Olden, J. D., M. K. Joy, et al. (2006). "Rediscovering the species in community-wide predictive modeling." Ecological Applications 16(4): 1449-1460.

Park, Y. S., R. Cereghino, et al. (2003). "Applications of artificial neural networks for patterning and predicting aquatic insect species richness in running waters." Ecological Modelling 160(3): 265-280.

Paruelo, J. M. and F. Tomasel (1997). "Prediction of functional characteristics of ecosystems: A comparison of artificial neural networks and regression models." Ecological Modelling 98(2-3): 173-186.

Spitz, F. and S. Lek (1999). "Environmental impact prediction using neural network modelling. An example in wildlife damage." Journal of Applied Ecology 36(2): 317-326.

Worner, S. P. and M. Gevrey (2006). "Modelling global insect pest species assemblages to determine risk of invasion." Journal of Applied Ecology 43(5): 858-867.