Tunning machine learning algorithms for forestry modeling: a case study in the height-diameter relationship
DOI:
https://doi.org/10.5902/1980509828392Keywords:
Artificial intelligence, Grid search, Artificial neural networks, Cross validationAbstract
In the present study, four machine learning algorithms were applied in the task of modeling the height-diameter relationship of Pinus taeda L. stands at different ages. Hundreds of parameter combinations were tested for the k-nearest neighbors, random forests, support vector machines, and artificial neural networks algorithms. In order to select the best model for each algorithm, the grid search and the k-fold cross validation methods were applied. The selected models were used to predict the total height of individuals in an independent data set, and the results were compared to those obtained by linear regression models. The machine learning models presented similar statistical indicators to the linear regression models. However, they had less biased dispersion of residues, especially in the stratified analysis by age. The support vector machine and the artificial neural network were the most satisfactory models in precision and dispersion of residues.Downloads
References
ABDOLLAHNEJAD, A. et al. Prediction of Dominant Forest Tree Species Using QuickBird and Environmental Data. Forests, [s.l.], v. 8, n. 2, p.42-60, fev. 2017.
BARROS, D. A. et al. Comportamento de modelos hipsométricos tradicionais e genéricos para plantações de Pinus oocarpa em diferentes tratamentos. Boletim de Pesquisa Florestal, Colombo, n. 45, p. 03-28, jul./dez. 2002.
BERGSTRA, J.; BENGIO, Y. Random search for hyper-parameter optimization. The Journal Of Machine Learning Research, Montréal, Qc, Canada, v. 13, n. 1, p.281-305, jan. 2012.
BINOTI, D.H.B.; BINOTI, M.L.M.S.; LEITE, H.G. Configuração de Redes Neurais Artificiais para Estimação do Volume de Árvores. Revista Ciência da Madeira, Pelotas, v. 5, n. 1, p.58-67, 31 maio 2014.
BLANCHETTE, D. et al. Predicting wood fiber attributes using local-scale metrics from terrestrial LiDAR data: A case study of Newfoundland conifer species. Forest Ecology and Management, [s.l.], v. 347, p.116-129, jul. 2015.
BREIMAN, L. Random Forests. Machine Learning, [s.l.], v. 45, n. 1, p.05-32, out. 2001.
CANDEL, A. et al. Deep Learning with H2O. Disponível em: http://docs.h2o.ai/h2o/latest-stable/h2o-docs/booklets/DeepLearningBooklet.pdf. Acesso em: 02 jul. 2017.
CORDEIRO, M. A., et al. Estimativa do volume de Acacia mangium utilizando técnicas de redes neurais artificiais e máquinas vetor de suporte. Pesquisa Florestal Brasileira, Colombo, v. 35, n. 83, 255-261, set. 2015.
CRISTIANINI, N.; SHAWE-TAYLOR, J. An introduction to support vector machines and other kernel-based learning methods. New York: Cambridge University Press, 2000.
FACELI, K.; LORENA, A. C.; GAMA, J.; CARVALHO, A. C. P. L. F. Inteligência artificial: uma abordagem de aprendizado de máquina. Rio de Janeiro: LTC, 2011. 378p.
GORGENS, E. B. et al. Influência da arquitetura na estimativa de volume de árvores individuais por meio de redes neurais artificiais. Revista Árvore, Viçosa-MG, v. 38, n. 2, p.289-295, abr. 2014.
GORGENS, E. B.; MONTAGHI, A.; RODRIGUEZ, L. C. E. A performance comparison of machine learning methods to estimate the fast-growing forest plantation yield based on laser scanning metrics. Computers and Electronics in Agriculture [s.l.], v. 116, p. 221-227, 2015.
HAARA, A.; MALTAMO, M.; TOKOLA, T. The K‐nearest‐neighbour method for estimating basal‐area diameter distribution. Scandinavian Journal Of Forest Research, Uppsala, v. 12, n. 2, p.200-208, maio, 1997. HECHENBICHLER, K; SCHLIEP, K. Weighted k-Nearest-Neighbor Techniques and Ordinal Classification. Discussion paper 399, Ludwig-Maximilians University Munich, Munich, 2004. Disponível em: https://epub.ub.uni-muenchen.de/1769/1/paper_399.pdf. Acesso em: 20 jun. 2017.
MARTINS, E. R. et al. Configuração de redes neurais artificiais para estimação da altura total de árvores de eucalipto. Revista Brasileira de Ciências Agrárias (Agrária), Recife, v. 11, n. 2, p. 117-123, 2016.
ÖZÇELIK, R. et al. Estimating Crimean juniper tree height using nonlinear regression and artificial neural network models. Forest Ecology And Management, [s.l.], v. 306, p.52-60, out. 2013.
R CORE TEAM. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. 2016. Disponível em: <http://www.R-project.org/> Acesso em: 04 ago. 2017.
ROBINSON, A. P.; LANE, S. E.; THÉRIEN, G. Fitting forestry models using generalized additive models: a taper model example. Canadian Journal Of Forest Research, [s.l.], v. 41, n. 10, p.1909-1916, out. 2011.
SANQUETTA, C. R. et al. On the use of data mining for estimating carbon storage in the trees. Carbon Balance And Management, [s.l.], v. 8, n. 1, p.6-14, 2013.
SHALEV-SHWARTZ, S.; BEN-DAVID, S. Understanding Machine Learning: From Theory to Algorithms. New York: Cambridge University Press, 1. ed., 2014.
SMOLA, A. Regression Estimation with Support Vector Learning Machines. 1996. 78 p. Master’s thesis (Physics) - Technische Universitat At Munchen, Munchen, 1996.
TOMMOLA, M., et al. Estimating the characteristics of a marked stand using k-nearest-neighbour regression. Journal of Forest Engineering, [s.l], v. 10, p.75-81, 1999.
VAPNIK, V N. The nature of statistical learning theory. 2. ed. New York: Springer-Verlag, 2000. 314 p.
VENDRUSCOLO et al. Height prediction of Tectona grandis trees by mixed effects modelling and artificial neural networks. International Journal of Current Research, [s.l.], v. 8, n. 12, p.43189-43195, dez. 2016.
