The problem of automatic classification of facies was addressed using the Fast Independent Component Analysis (FastICA) of a data set of geophysical well logs of the Namorado Field, Campos Basin, Brazil, followed by a k-nearest neighbor (k-NN) classification. The goal of an automatic classification of facies is to produce spatial models of facies that assist the geological characterization of petroleum reservoirs. The FastICA technique provides a new data set that has the most stable and less Gaussian distribution possible. The k-NN classifies this new data set according to its characteristics. The previous application of FastICA improves the accuracy of the k-NN automatic classification and it also provides better results in comparison with the automatic classification by means of the Principal Component Analysis (PCA). © 2015 Sociedade Brasileira de Geofísica.331119126Barboza, E.G., (2005) Análise Estratigráfica Do Campo De Namorado Com Base Na interpretação Sísmica Tridimensional, 230p. , Doctorate Thesis – UFRGS, BrazilCardoso, J.F., Souloumiac, A., Blind beamforming for non Gaussian signals (1993) IEE Proceedings-F, 140 (6), pp. 362-370Casey, M.A., Method for extracting features from a mixture of signals (2001) Mitsubishi Electric Research Laboratories, , Inc., Cambridge, MA, U.S. Patent n. 6,321,200Comon, P., Independent Component Analysis: A new concept? (1994) Signal Processing, 36 (3), pp. 287-314Cover, T.M., Hart, P.E., Nearest neighbor pattern classification (1967) IEEE Transactions on Information Theory, 13 (1), pp. 21-27Dasarathy, B.V., (1991) Nearest Neighbor (NN) Norms: NN Pattern Classification Techniques, pp. 211-215. , IEEE Computer Society Press, Los Alamitos, Calif. WashingtonFlexa, R.T., Rade, A., Carrasquilla, A., Identificação de litotipos nos perfis de poço do Campo de Namorado (Bacia de Campos, Brasil) e do Lago Maracaibo (Venezuela) aplicando estatística multivariada (2004) Revista Brasileira De Geociências, 34 (4), pp. 571-578Françamm, Análise do uso da terra no município de Viçosa-MG mediado por classificações supervisionadas com Redes Neurais Artificiais e Maxver (2009) Revista Brasileira De Geografia Física, 2 (3), pp. 92-101Giannakopoulos, X., Karhunen, J., Oja, E., A Comparison of neural ICA algorithms using Real-world data (1999) International Joint Conference on Neural Networks, 2, pp. 888-893. , Washington, DCHyvärinen, A., Fast and Robust Fixed-Point Algorithms for IndependentComponentAnalysis (1999) IEEE Transactions on Neural Networks, 10 (3), pp. 626-634Hyvärinen, A., Oja, E., A Fast fixed-point algorithm for Independent Component Analysis (1997) Neural Computation, 9 (7), pp. 1483-1492Hyvärinen, A., Karhunen, J., Oja, E., (2001) Independent Component Analysis, 504p. , Wiley and Sons, New YorkKuhn, H.W., Tucker, A.W., Nonlinear programming (1951) Proceedings of Berkeleysymposium, 2, pp. 481-492. , Berkeley: University of California PressLeite, E.P., Souza Filho, C.R., TEXTNN – A MATLAB program for textural classification using neural networks (2009) Computers& Geosciences, 35 (10), pp. 2084-2094Macleod, A.J., A generalization of Newton-Raphson (1984) Int. J. Math. Educ. Sci. Technol., 15 (1), pp. 117-120Macqueen, J.B., Some Methods for classification and Analysis of Multivariate Observations (1967) Proceedings of Berkeley Symposium on Mathematical Statistics and Probability, 1, pp. 281-297. , 5., Berkeley, University of California PressMarchini, J.L., Heaton, C., Ripley, B.D., (2009) Fastica Algorithms to Perform ICA and Projection Pursuit, 1, pp. 1-11. , R package versionMurata, N., Ikeda, S., Ziehe, A., An approach to blind source separation based on temporal structure of speech signals (2001) Neurocomputing, 41, pp. 1-24Rosa, H., Suslick, S.B., Vidal, A.C., Sakai, G.K., Caracterização de eletrofácies por meio de ferramentas estatísticas multivariadas (2008) Rem: Rev. Esc. Minas, 61 (4), pp. 415-422Sancevero, S.S., Remacre, A.Z., Vidal, A.C., Portugal, R.S., Aplicação de técnicas de estatística multivariada na definição da litologia a partir de perfis geofísicos de poços (2008) Revista Brasileira De Geociências, 38 (1), pp. 61-74Toussaint, G.T., Geometric proximity graphs for improving nearest neighbor methods in instance-based learning and data mining (2005) International Journal of Computational Geometry and Applications, 15 (2), pp. 101-150Vidal, A.C., Sancevero, S.S., Remacre, A.Z., Costanzo, C.P., Modelagem Geoestatística 3D da Impedância Acústica para a Caracterização do Campo de Namorado (2007) Brazilian Journal of Geophysics, 25 (3), pp. 295-305Zhang, K., Chan, L., Dimension reduction as a deflation method in ICA (2006) Signal Processing Letters, IEEE, 13 (1), pp. 45-48