Machine learning of poorly predictable ecological data Export

@article{citeulike:3851782, abstract = {This paper reports on research using a variety of machine learning techniques to a difficult modelling problem, the spatial distribution of an endangered Australian marsupial, the southern brown bandicoot (Isoodon obesulus). Four learning techniques – decision trees/rules, neural networks, support vector machines and genetic programming – were applied to the problem. Support vector and neural network approaches gave marginally better predictivity, but in the context of low overall accuracy, decision trees and genetic programming gave more useful results because of the human comprehensibility of their models.}, author = {Shan, Y. and Paull, D. and Mckay, R.}, citeulike-article-id = {3851782}, citeulike-linkout-0 = {http://dx.doi.org/10.1016/j.ecolmodel.2005.11.015}, citeulike-linkout-1 = {http://linkinghub.elsevier.com/retrieve/pii/S0304380005005843}, day = {15}, doi = {10.1016/j.ecolmodel.2005.11.015}, issn = {03043800}, journal = {Ecological Modelling}, keywords = {ann, cart, meachine-learning, sdm, svm}, month = {May}, number = {1-2}, pages = {129--138}, posted-at = {2009-01-06 05:08:13}, priority = {2}, title = {Machine learning of poorly predictable ecological data}, url = {http://dx.doi.org/10.1016/j.ecolmodel.2005.11.015}, volume = {195}, year = {2006} }

TY - JOUR ID - citeulike:3851782 L3 - citeulike-article-id:3851782 N2 - This paper reports on research using a variety of machine learning techniques to a difficult modelling problem, the spatial distribution of an endangered Australian marsupial, the southern brown bandicoot (Isoodon obesulus). Four learning techniques – decision trees/rules, neural networks, support vector machines and genetic programming – were applied to the problem. Support vector and neural network approaches gave marginally better predictivity, but in the context of low overall accuracy, decision trees and genetic programming gave more useful results because of the human comprehensibility of their models. IS - 1-2 JF - Ecological Modelling SN - 03043800 EP - 138 TI - Machine learning of poorly predictable ecological data VL - 195 SP - 129 KW - ann KW - cart KW - meachine-learning KW - sdm KW - svm AU - Shan, Y AU - Paull, D AU - Mckay, R PY - 2006/05/15/ UR - http://dx.doi.org/10.1016/j.ecolmodel.2005.11.015 ER -