Melhorando a Performance do Algoritmo Naive Bayes para Regressão Através da Combinação de Atributos

  • Aloísio Carlos de Pina UFRJ
  • Gerson Zaverucha UFRJ

Resumo


O algoritmo Naive Bayes para Regressão (NBR) usa a metodologia do Naive Bayes para tarefas de predição numéricas. A principal razão de sua pobre performance é a suposição de independência. Embora muitas pesquisas recentes tentem melhorar a performance do Naive Bayes pelo relaxamento da suposição de independência, nenhuma delas pode ser aplicada diretamente a problemas de regressão. O objetivo deste trabalho é apresentar uma nova abordagem para melhorar os resultados do algoritmo NBR, combinando atributos por meio de algoritmos de regressão auxiliares.

Referências

Blake, C.L. e Merz, C.J. (2005) UCI Repository of Machine Learning Databases.

Machine-readable data repository, Department of Information & Computer Science, University of California, Irvine. [link]

Dietterich, T.G. (1998) Approximate Statistical Tests for Comparing Supervised Classification Learning Algorithms. Neural Computation, 10, pp. 1895-1924.

Duda, R.O. e Hart, P.E. (1973) Pattern Classification and Scene Analysis. New York, NY: John Wiley & Sons.

Frank, E., Trigg, L., Holmes, G. e Witten, I.H. (2000) Naive Bayes for Regression. Machine Learning, 41, pp. 5-25.

Friedman, N. e Goldszmidt, M. (1996) Building classifiers using Bayesian networks. In Proceedings of the 13th National Conference on Artificial Intelligence, pp. 1277-1284. Portland, OR: AAAI Press.

Iba, W. e Langley, P. (1992) Induction of one-level decision trees. In Proceedings of the 9th International Conference on Machine Learning, pp. 233-240. Aberdeen, Scotland: Morgan Kaufmann Publishers, Inc.

Keogh, E. e Pazzani, M. (2002) Learning the Structure of Augmented Bayesian Classifiers. International Journal on Artificial Intelligence Tools, 11 (4), pp. 587-601. World Scientific Publishing Company.

Kohavi, R. (1994) Feature subset selection as search with probabilistic estimates. In Procedures of the AAAI Fall Symposium on Relevance, pp. 122-126. New Orleans, LA: AAAI Press.

Kohavi, R. e John, G.H. (1995) Automatic parameter selection by minimizing estimated error. In A. Prieditis and S. Russell (eds.), Proceedings of the 12th International Conference on Machine Learning, pp. 304-312. Tahoe City, CA: Morgan Kaufmann.

Kononenko, I. (1991) Semi-naive Bayesian classifier. In Proceedings of the Sixth European Working Session on Learning, pp. 206-219. Porto, Portugal: Springer-Verlag.

Mitchell, T.M. (1997) Machine Learning. New York, NY: McGraw-Hill.

Pazzani, M. (1996) Searching for dependencies in Bayesian classifiers. In D. Fisher and H.J. Lenz (eds.), Learning from data: Artificial intelligence and statistics V, pp. 239-248. New York, NY: Springer-Verlag.

Wang, Z., Webb, G.I. e Zheng, F. (2003) Adjusting Dependence Relations for Semi-Lazy TAN Classifiers. In Proceedings of the 16th Australian Conference on Artificial Intelligence, pp. 453-465. Perth, Australia: Springer.

Wang, Z., Webb, G.I. e Zheng, F. (2004) Selective Augmented Bayesian Network Classifiers Based on Rough Set Theory. In Proceedings of the 8th Pacific-Asia Conference on Advances in Knowledge Discovery and Data Mining, pp. 319-328. Sydney, Australia: Springer.

Webb, G. I., Boughton, J. e Wang, Z. (2005) Not So Naive Bayes: Aggregating One-Dependence Estimators. Machine Learning, 58 (1), pp. 5-24.

Witten, I.H. e Frank, E. (2005) Data Mining: Practical Machine Learning Tools and Techniques, 2nd edition. San Francisco, CA: Morgan Kaufmann.

Zhang, H. e Ling, C.X. (2001) An Improved Learning Algorithm for Augmented Naive Bayes. In Proceeding of the 5th Pacific-Asia Conference on Advances in Knowledge Discovery and Data Mining, pp. 581-586. Hong Kong, China: Springer.

Zheng, Z. e Webb, G.I. (2000) Lazy Learning of Bayesian Rules. Machine Learning, 41 (1), pp. 53-84.
Publicado
30/06/2007
PINA, Aloísio Carlos de; ZAVERUCHA, Gerson. Melhorando a Performance do Algoritmo Naive Bayes para Regressão Através da Combinação de Atributos. In: ENCONTRO NACIONAL DE INTELIGÊNCIA ARTIFICIAL E COMPUTACIONAL (ENIAC), 6. , 2007, Rio de Janeiro/RJ. Anais [...]. Porto Alegre: Sociedade Brasileira de Computação, 2007 . p. 1529-1537. ISSN 2763-9061.

Artigos mais lidos do(s) mesmo(s) autor(es)