Previsão de ocupação de anuros usando sensoriamento de variáveis ambientais e modelos Autocodificadores
Resumo
Neste trabalho apresentamos um modelo de previsão da distribuição geográfica de anuros baseado em uma Rede Neural Autocodificadora. O problema de previsão é tratado como uma tarefa de classificação One-Class, em que a localização conhecida dos indivı́duos é representada através de suas coordenadas espaciais, temporais e variáveis meteorológicas correlacionadas. O modelo mostrou alto grau de acuidade quando utilizado para prever a ocorrência de sapos da espécie Bufo Americanus no sul do Canadá. O método proposto é computacionalmente leve e pode ser acoplado a redes de sensores para monitoramento ambiental.
Palavras-chave:
Autoencoder Neural Network, geographic distribuition of species, enviromental features, anurans
Referências
AmphibiaWeb (2019). American toad. Disponı́vel em: https://amphibiaweb. org. Acesso em: 11 de março de 2019.
Bateman, B. L., VanDerWal, J., and Johnson, C. N. (2012). Nice weather for bettongs: using weather events, not climate means, in species distribution models. Ecography, 35(4):306–314.
Chen, G. H., Shah, D., et al. (2018). Explaining the success of nearest neighbor methods in prediction. Foundations and Trends in Machine Learning, 10(5-6):337–588.
FrogWatch (2018). Frogwatch: Enganging citizens in science. Disponı́vel em: https: //www.naturewatch.ca/frogwatch/. Acesso em: 13 de dezembro de 2018
Fukunaga, K. (2013). Introduction to statistical pattern recognition. Elsevier.
Géron, A. (2017). Hands-On Machine Learning with Scikit-Learn and TensorFlow. O’Reilly Media, 1st edition
Jarvie, S. and Svenning, J. C. (2018). Using species distribution modelling to determine opportunities for trophic rewilding under future scenarios of climate change. Philo- sophical Transactions of the Royal Society B: Biological Sciences, 373(1761).
Khan, S. S. and Madden, M. G. (2014). One-class classification: taxonomy of study and review of techniques. The Knowledge Engineering Review, 29(3):345–374.
Lorena, L. H. N., Carvalho, A. C. P. L. F., and Lorena, A. C. (2015). Filter feature selec- tion for one-class classification. Journal of Intelligent & Robotic Systems, 80(1):227– 243.
Mazhelis, O. (2006). One-class classifiers: a review and analysis of suitability in the con- text of mobile-masquerader detection. South African Computer Journal, 2006(36):29– 48.
Phillips, S. J., Anderson, R. P., and Schapire, R. E. (2006). Maximum entropy modeling of species geographic distributions. Ecological modelling, 190(3-4):231–259
Phillips, S. J., Dudı́k, M., and Schapire, R. E. (2004). A maximum entropy approach to species distribution modeling. In Proceedings of the twenty-first international confe- rence on Machine learning, page 83. ACM.
Society, C. H. (2018). American toad. Disponı́vel em: http:// canadianherpetology.ca/species/species_page.html?cname= American%20Toad. Acesso em: 13 de março de 2019.
Theodoridis, S. and Koutroumbas, K. (2008). Pattern Recognition, Fourth Edition. Aca- demic Press, Inc., 4th edition.
WeatherUnderground (2018). Api: Weather underground. Disponı́vel em: https:// www.wunderground.com/weather/api. Acesso em: 15 de dezembro de 2018.
Bateman, B. L., VanDerWal, J., and Johnson, C. N. (2012). Nice weather for bettongs: using weather events, not climate means, in species distribution models. Ecography, 35(4):306–314.
Chen, G. H., Shah, D., et al. (2018). Explaining the success of nearest neighbor methods in prediction. Foundations and Trends in Machine Learning, 10(5-6):337–588.
FrogWatch (2018). Frogwatch: Enganging citizens in science. Disponı́vel em: https: //www.naturewatch.ca/frogwatch/. Acesso em: 13 de dezembro de 2018
Fukunaga, K. (2013). Introduction to statistical pattern recognition. Elsevier.
Géron, A. (2017). Hands-On Machine Learning with Scikit-Learn and TensorFlow. O’Reilly Media, 1st edition
Jarvie, S. and Svenning, J. C. (2018). Using species distribution modelling to determine opportunities for trophic rewilding under future scenarios of climate change. Philo- sophical Transactions of the Royal Society B: Biological Sciences, 373(1761).
Khan, S. S. and Madden, M. G. (2014). One-class classification: taxonomy of study and review of techniques. The Knowledge Engineering Review, 29(3):345–374.
Lorena, L. H. N., Carvalho, A. C. P. L. F., and Lorena, A. C. (2015). Filter feature selec- tion for one-class classification. Journal of Intelligent & Robotic Systems, 80(1):227– 243.
Mazhelis, O. (2006). One-class classifiers: a review and analysis of suitability in the con- text of mobile-masquerader detection. South African Computer Journal, 2006(36):29– 48.
Phillips, S. J., Anderson, R. P., and Schapire, R. E. (2006). Maximum entropy modeling of species geographic distributions. Ecological modelling, 190(3-4):231–259
Phillips, S. J., Dudı́k, M., and Schapire, R. E. (2004). A maximum entropy approach to species distribution modeling. In Proceedings of the twenty-first international confe- rence on Machine learning, page 83. ACM.
Society, C. H. (2018). American toad. Disponı́vel em: http:// canadianherpetology.ca/species/species_page.html?cname= American%20Toad. Acesso em: 13 de março de 2019.
Theodoridis, S. and Koutroumbas, K. (2008). Pattern Recognition, Fourth Edition. Aca- demic Press, Inc., 4th edition.
WeatherUnderground (2018). Api: Weather underground. Disponı́vel em: https:// www.wunderground.com/weather/api. Acesso em: 15 de dezembro de 2018.
Publicado
12/07/2019
Como Citar
SILVA, Nabson; COLONNA, Juan ; CRISTO, Marco .
Previsão de ocupação de anuros usando sensoriamento de variáveis ambientais e modelos Autocodificadores. In: SIMPÓSIO BRASILEIRO DE COMPUTAÇÃO UBÍQUA E PERVASIVA (SBCUP), 11. , 2019, Belém.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2019
.
ISSN 2595-6183.
DOI: https://doi.org/10.5753/sbcup.2019.6583.
