HR: 1330h
AN: NB33M-10    [Abstracts]
TI: Comparing ANNs, EAs, and Trees: a basic machine-learning approach to predictive environmental models.
AU: * Williams, J
EM: jwilli@lamar.colostate.edu
AF: Colorado State University, Dept. of Biology, fort collins, CO 80526 United States
AU: Poff, N
EM: poff@lamar.colostate.edu
AF: Colorado State University, Dept. of Biology, fort collins, CO 80526 United States
AB: Machine learning techniques for ecological applications or "eco-informatics" are becoming increasingly useful and accessible for ecologists. We evaluated the predictive ability of three commercially available (i.e. user-friendly) software packages for artificial neural networks (ANNs), evolutionary algorithms (EAs), and classification/regression trees (Trees). We analyzed fish and habitat data for streams in the mid-Atlantic region of the U.S., which was collected by the U.S. Environmental Protection Agency (EPA). The data includes over 200 environmental descriptors summarizing watershed, stream, and water chemistry characteristics in addition to derived fish community metrics (i.e. richness, IBI scores, % exotics). In our analysis we predicted individual species presence/absence and fish community metrics as a function of these local and regional scale habitat variables. Predictive ability is evaluated with independent validation data. These approaches could prove especially useful for conservation or management applications where ecologists seek to utilize the most comprehensive data to make predictions at various scales. By employing "user-friendly" software we hope to show that ecologists, without extensive knowledge of computational science, can benefit from these techniques by extracting more information about complex ecosystems. Relative strengths and weaknesses of these three approaches are compared and recommendations for their use in conservation applications are presented.
DE: 9903 NABS Student Award Methodology
SC: North American Benthological Society [NB]
MN: 2005 Joint Assembly