Trabalhos Futuros

leave-one-out. Além disso, o desempenho do classificador modal foi comparado com o desempenho de um classificador para dados modais (aqui chamado ID-KNN) que foi estudado nesse trabalho.

Os resultados obtidos para a taxa de erro e o tempo de execução mostraram que, em média, o classificador ID-KNN é superior ao classificador modal considerando diferentes conjuntos de dados intervalares sintéticos, diferentes funções de dissimilaridade e diferentes tamanhos de vizinhança para o classificador ID-KNN. O estudo do conjunto de dados de temperatura considerando também diferentes funções de dissimilaridade e diferentes tamanhos de vizinhança para o classificador ID-KNN, revelou que esses classificadores apresentaram desempenhos similares sendo o classificador modal superior, em termos da taxa de erro.

É importante salientar que no período de estudo e implementação do classificador modal um artigo foi publicado [Silva et al., 2006] no ICONIP-2006 (13th International Conference on Neural Information Processing) que é uma importante conferência internacional anual para explorar e trocar idéias em redes neural e em disciplinas relacionadas.

A contribuição principal do trabalho é a comparação de várias medidas de dissimilaridades e de seus componentes internos (pesos) e seus suportes e selecionar a melhor medida para um problema na análise de dados. Escolhendo e aplicando funções diferentes de dissimilaridade à mesma base de dados, é possível descobrir a função a mais indicada de dissimilaridade a ser aplicada a um problema específico.

6.1. Trabalhos Futuros

Com relação a continuidade deste trabalho, pode-se mencionar as seguintes extensões:

I. Fazer um estudo comparativo do classificador simbólico modal para dados do tipo intervalo com outras técnicas de classificação supervisionada existentes com por exemplo Redes Neurais Artificiais. II. Adaptar e testar outras distâncias para os dados simbólicos modais bem

como utilizar outras bases de dados reais para um melhor validação do método.

Apêndice A

Arquivo “temperatura.sds “

A seguir um exemplo de um arquivo no formato padrão de SODA’s que é o arquivo “temperatura.sds” que foi utilizado nos testes das aplicações com dados reais que é um conjunto contém 37 cidades, cada cidade é descrita por 12 variáveis do tipo intervalo que são mínimas e máximas de temperaturas em graus centígrados de 12 meses.

SODAS = ( CONTAINS = (

FILES, HEADER, INDIVIDUALS, VARIABLES, RECTANGLE_MATRIX ), FILE = ( procedure_name = "db2so" , version = "sans" , create_date = "" ), HEADER = ( title = "temperaturas" , sub_title = "h" , indiv_nb = 37 , var_nb = 13 , rules_nb = 0 , nb_var_set = 0 , nb_indiv_set = 0 , nb_var_nom = 0 , nb_var_cont = 0 , nb_var_text = 0 , nb_var_cont_symb = 12 , nb_var_nom_symb = 1 , nb_var_nom_mod = 0 , nb_na = 0 ,

Apêndice A nb_null = 0 , nb_nu = 0 , nb_hierarchies = 0 ), INDIVIDUALS = ( (0,"AA00", "Amssterdam" ), (1,"AA01", "Athens" ), (2,"AA02", "Bahrain" ), (3,"AA03", "Bombay" ), (4,"AA04", "Cairo" ), (5,"AA05", "Calcutta" ), (6,"AA06", "Colombo" ), (7,"AA07", "Copenhagen" ), (8,"AA08", "Dubal" ), (9,"AA09", "Frankfurt" ), (10,"AA10", "Geneva" ), (11,"AA11", "HongKong" ), (12,"AA12", "KulaLumpur" ), (13,"AA13", "Lisbon" ), (14,"AA14", "London" ), (15,"AA15", "Madras" ), (16,"AA16", "Madrid" ), (17,"AA17", "Manila" ), (18,"AA18", "Mauritius" ), (19,"AA19", "MexicoCity" ), (20,"AA20", "Moscow" ), (21,"AA21", "Munich" ), (22,"AA22", "Nairobi" ), (23,"AA23", "NewDelhi" ), (24,"AA24", "NewYork" ), (25,"AA25", "Paris" ), (26,"AA26", "Rome" ), (27,"AA27", "SanFrancisco" ),

Apêndice A (28,"AA28", "Seoul" ), (29,"AA29", "Singapore" ), (30,"AA30", "Stockholm" ), (31,"AA31", "Sydney" ), (32,"AA32", "Tehran" ), (33,"AA33", "Tokyo" ), (34,"AA34", "Toronto" ), (35,"AA35", "Vienna" ), (36,"AA36", "Zurich" ) ), VARIABLES = (

(1 ,inter_cont ,"" ,"AB00" ,"JAN" ,0, 0, -13, 31), (2 ,inter_cont ,"" ,"AC00" ,"FEB" ,0, 0, -12, 32), (3 ,inter_cont ,"" ,"AD00" ,"MAR" ,0, 0, -8, 34), (4 ,inter_cont ,"" ,"AE00" ,"APR" ,0, 0, -2, 36), (5 ,inter_cont ,"" ,"AF00" ,"MAY" ,0, 0, -8, 40), (6 ,inter_cont ,"" ,"AG00" ,"JUN" ,0, 0, 5, 39), (7 ,inter_cont ,"" ,"AH00" ,"JUL" ,0, 0, 8, 39), (8 ,inter_cont ,"" ,"AI00" ,"AUG" ,0, 0, 8, 40), (9 ,inter_cont ,"" ,"AJ00" ,"SEPT" ,0, 0, 5, 37), (10 ,inter_cont ,"" ,"AK00" ,"OCT" ,0, 0, 0, 34), (11 ,inter_cont ,"" ,"AL00" ,"NOV" ,0, 0, -3, 32), (12 ,inter_cont ,"" ,"AM00" ,"DEC" ,0, 0, -11, 31), (13 ,nominal ,"" ,"AE00" ,"Edibility" ,0, 0 ,4, ( (1 ,"AE01" ,"U" ,0), (2 ,"AE02" ,"U" ,0), (3 ,"AE02" ,"U" ,0), (4 ,"AE03" ,"T" ,0 ) ) ) ),

Apêndice A - 79 - RECTANGLE_MATRIX = ( (( -4 : 4 ), ( -5 : 3 ), ( 2 : 12 ), ( 5 : 15 ), ( 7 : 17 ), ( 10 : 20 ), ( 10 : 20 ), ( 12 : 23 ), ( 10 : 20 ), ( 5 : 15 ), ( 1 : 10 ), ( -1 : 4 ), 2), (( 6 : 12 ), ( 6 : 12 ), ( 8 : 16 ), ( 11 : 19 ), ( 16 : 25 ), ( 19 : 29 ), ( 22 : 32 ), ( 22 : 32 ), ( 19 : 28 ), ( 16 : 23 ), ( 11 : 18 ), ( 8 : 14 ), 2), (( 13 : 19 ), ( 14 : 19 ), ( 17 : 23 ), ( 21 : 27 ), ( 25 : 32 ), ( 28 : 34 ), ( 29 : 36 ), ( 30 : 36 ), ( 28 : 34 ), ( 24 : 31 ), ( 20 : 26 ), ( 15 : 21 ), 1), (( 19 : 28 ), ( 19 : 28 ), ( 22 : 30 ), ( 24 : 32 ), ( 27 : 33 ), ( 26 : 32 ), ( 25 : 30 ), ( 25 : 30 ), ( 24 : 30 ), ( 24 : 32 ), ( 23 : 32 ), ( 20 : 30 ), 1), (( 8 : 20 ), ( 9 : 22 ), ( 11 : 25 ), ( 14 : 29 ), ( 17 : 33 ), ( 20 : 35 ), ( 22 : 36 ), ( 22 : 35 ), ( 20 : 33 ), ( 18 : 31 ), ( 14 : 26 ), ( 10 : 20 ), 1), (( 13 : 27 ), ( 16 : 29 ), ( 21 : 34 ), ( 24 : 36 ), ( 26 : 36 ), ( 26 : 33 ), ( 26 : 32 ), ( 26 : 32 ), ( 26 : 32 ), ( 24 : 32 ), ( 18 : 29 ), ( 13 : 26 ), 1), (( 22 : 30 ), ( 22 : 30 ), ( 23 : 31 ), ( 24 : 31 ), ( 25 : 31 ), ( 25 : 30 ), ( 25 : 29 ), ( 25 : 29 ), ( 25 : 30 ), ( 24 : 29 ), ( 23 : 29 ), ( 22 : 30 ), 1), (( -2 : 2 ), ( -3 : 2 ), ( -1 : 5 ), ( 3 : 10 ), ( 8 : 16 ), ( 11 : 20 ), ( 14 : 22 ), ( 14 : 21 ), ( 11 : 18 ), ( 7 : 12 ), ( 3 : 7 ), ( 1 : 4 ), 2), (( 13 : 23 ), ( 14 : 24 ), ( 17 : 28 ), ( 19 : 31 ), ( 22 : 34 ), ( 25 : 36 ), ( 28 : 39 ), ( 28 : 39 ), ( 25 : 37 ), ( 21 : 34 ), ( 17 : 30 ), ( 14 : 26 ), 1), (( -10 : 9 ), ( -8 : 10 ), ( -4 : 17 ), ( 0 : 24 ), ( 3 : 27 ), ( 7 : 30 ), ( 8 : 32 ), ( 8 : 31 ), ( 5 : 27 ), ( 0 : 22 ), ( -3 : 14 ), ( -8 : 10 ), 2), (( -3 : 5 ), ( -6 : 6 ), ( 3 : 9 ), ( 7 : 13 ), ( 10 : 17 ), ( 15 : 17 ), ( 16 : 24 ), ( 16 : 23 ), ( 11 : 19 ), ( 6 : 13 ), ( 3 : 8 ), ( -2 : 6 ), 2), (( 13 : 17 ), ( 12 : 16 ), ( 15 : 19 ), ( 19 : 23 ), ( 22 : 27 ), ( 25 : 29 ), ( 25 : 30 ), ( 25 : 30 ), ( 25 : 29 ), ( 22 : 27 ), ( 18 : 23 ), ( 14 : 19 ), 1), (( 22 : 31 ), ( 23 : 32 ), ( 23 : 33 ), ( 23 : 33 ), ( 23 : 32 ), ( 23 : 32 ), ( 23 : 31 ), ( 23 : 32 ), ( 23 : 32 ), ( 23 : 31 ), ( 23 : 31 ), ( 23 : 31 ), 1), (( 8 : 13 ), ( 8 : 14 ), ( 9 : 16 ), ( 11 : 18 ), ( 13 : 21 ), ( 16 : 24 ), ( 17 : 26 ), ( 18 : 27 ), ( 17 : 24 ), ( 14 : 21 ), ( 11 : 17 ), ( 8 : 14 ), 2), (( 2 : 6 ), ( 2 : 7 ), ( 3 : 10 ), ( 5 : 13 ), ( 8 : 17 ), ( 11 : 20 ), ( 13 : 22 ), ( 13 : 21 ), ( 11 : 19 ), ( 8 : 14 ), ( 5 : 10 ), ( 3 : 7 ), 2), (( 20 : 30 ), ( 20 : 31 ), ( 22 : 33 ), ( 26 : 35 ), ( 28 : 39 ), ( 27 : 38 ), ( 26 : 36 ), ( 26 : 35 ), ( 25 : 34 ), ( 24 : 32 ), ( 22 : 30 ), ( 21 : 29 ), 1), (( 1 : 9 ), ( 1 : 12 ), ( 3 : 16 ), ( 6 : 19 ), ( 9 : 24 ), ( 13 : 29 ), ( 16 : 34 ), ( 16 : 33 ), ( 13 : 28 ), ( 8 : 20 ), ( 4 : 14 ), ( 1 : 9 ), 2), (( 21 : 27 ), ( 22 : 27 ), ( 24 : 29 ), ( 24 : 31 ), ( 25 : 31 ), ( 25 : 31 ), ( 23 : 29 ), ( 24 : 28 ), ( 25 : 28 ), ( 24 : 29 ), ( 22 : 28 ), ( 22 : 27 ), 1), (( 22 : 28 ), ( 22 : 29 ), ( 22 : 29 ), ( 21 : 28 ), ( 19 : 25 ), ( 18 : 24 ), ( 17 : 23 ), ( 17 : 23 ), ( 17 : 24 ), ( 18 : 25 ), ( 19 : 27 ), ( 21 : 28 ), 3),

Apêndice A (( 6 : 22 ), ( 15 : 23 ), ( 17 : 25 ), ( 18 : 27 ), ( 18 : 27 ), ( 18 : 27 ), ( 18 : 27 ), ( 18 : 26 ), ( 18 : 26 ), ( 16 : 25 ), ( 14 : 25 ), ( 8 : 23 ), 1), (( -13 : -6 ), ( -12 : -15 ), ( -8 : 0 ), ( 0 : 8 ), ( 7 : 18 ), ( 11 : 23 ), ( 13 : 24 ), ( 11 : 22 ), ( 6 : 16 ), ( 1 : 8 ), ( -5 : 0 ), ( -11 : -5 ), 2), (( -6 : 1 ), ( -5 : 3 ), ( -2 : 9 ), ( 3 : 14 ), ( 7 : 18 ), ( 10 : 21 ), ( 12 : 23 ), ( 11 : 23 ), ( 8 : 20 ), ( 4 : 13 ), ( 0 : 7 ), ( -4 : 2 ), 2), (( 12 : 25 ), ( 13 : 26 ), ( 14 : 25 ), ( 14 : 24 ), ( 13 : 22 ), ( 12 : 21 ), ( 11 : 21 ), (11 : 21 ), ( 11 : 24 ), ( 13 : 24 ), ( 13 : 23 ), ( 13 : 23 ), 1), (( 6 : 21 ), ( 10 : 24 ), ( 14 : 29 ), ( 20 : 36 ), ( 26 : 40 ), ( 28 : 39 ), ( 27 : 35 ), ( 26 : 34 ), ( 24 : 34 ), ( 18 : 34 ), ( 11 : 28 ), ( 7 : 23 ), 1), (( -2 : 4 ), ( -3 : 4 ), ( 1 : 9 ), ( 6 : 15 ), ( 12 : 22 ), ( 17 : 27 ), ( 21 : 29 ), ( 20 : 28 ), ( 16 : 24 ), ( 11 : 19 ), ( 5 : 12 ), ( -2 : 6 ), 2), (( 1 : 7 ), ( 1 : 7 ), ( 2 : 12 ), ( 5 : 16 ), ( 8 : 19 ), ( 12 : 22 ), ( 14 : 24 ), ( 13 : 24 ), ( 11 : 21 ), ( 7 : 16 ), ( 4 : 10 ), ( 1 : 6 ), 2), (( 4 : 11 ), ( 5 : 13 ), ( 7 : 16 ), ( 10 : 19 ), ( 13 : 23 ), ( 17 : 28 ), ( 20 : 31 ), ( 20 : 31 ), ( 17 : 27 ), ( 13 : 21 ), ( 9 : 16 ), ( 5 : 12 ), 2), (( 6 : 13 ), ( 6 : 14 ), ( 7 : 17 ), ( 8 : 18 ), ( 10 : 19 ), ( 11 : 21 ), ( 12 : 22 ), ( 12 : 22 ), ( 12 : 23 ), ( 11 : 22 ), ( 8 : 18 ), ( 6 : 14 ), 2), (( 0 : 7 ), ( 1 : 6 ), ( 1 : 8 ), ( 6 : 16 ), ( 12 : 22 ), ( 16 : 25 ), ( 18 : 31 ), ( 16 : 30 ), ( 9 : 28 ), ( 3 : 24 ), ( 7 : 19 ), ( 1 : 8 ), 2), (( 23 : 30 ), ( 23 : 30 ), ( 24 : 31 ), ( 24 : 31 ), ( 24 : 30 ), ( 25 : 30 ), ( 25 : 30 ), ( 25 : 30 ), ( 24 : 30 ), ( 24 : 30 ), ( 24 : 30 ), ( 23 : 30 ), 1), (( -9 : -5 ), ( -9 : -6 ), ( -4 : -2 ), ( 1 : 8 ), ( 6 : 15 ), ( 11 : 19 ), ( 14 : 22 ), ( 13 : 20 ), ( 9 : 15 ), ( 5 : 9 ), ( 1 : 4 ), ( -2 : 2 ), 2), (( 20 : 30 ), ( 20 : 30 ), ( 18 : 26 ), ( 16 : 23 ), ( 12 : 20 ), ( 5 : 17 ), ( 8 : 16 ), ( 9 : 17 ), ( 11 : 20 ), ( 13 : 22 ), ( 16 : 26 ), ( 20 : 30 ), 1), (( 0 : 5 ), ( 5 : 8 ), ( 10 : 15 ), ( 15 : 18 ), ( 20 : 25 ), ( 28 : 30 ), ( 36 : 38 ), ( 38 : 40 ), ( 29 : 30 ), ( 18 : 20), ( 9 : 12 ), ( -5 : 0 ), 4), (( 0 : 9 ), ( 0 : 10 ), ( 3 : 13 ), ( 9 : 18 ), ( 14 : 23 ), ( 18 : 25 ), ( 22 : 29 ), ( 23 : 31 ), ( 20 : 27 ), ( 13 : 21 ), ( 8 : 16 ), ( 2 : 12 ), 2), (( -8 : -1 ), ( -8 : -1 ), ( -4 : 4 ), ( -2 : 11 ), ( -8 : 18 ), ( 13 : 24 ), ( 16 : 27 ), ( 16 : 26 ), ( 12 : 22 ), ( 6 : 14 ), ( -1 : 17 ), ( -5 : 1 ), 2), (( -2 : 1 ), ( -1 : 3 ), ( 1 : 8 ), ( 5 : 14 ), ( 10 : 19 ), ( 13 : 22 ), ( 15 : 24 ), ( 14 : 23 ), ( 11 : 19 ), ( 7 : 13 ), ( 2 : 7 ), ( 1 : 3 ), 2), (( -11 : 9 ), ( -8 : 15 ), ( -7 : 18 ), ( -1 : 21 ), ( 2 : 27 ), ( 6 : 30 ), ( 10 : 31 ), ( 8 : 25 ), ( 5 : 23 ), ( 3 : 22 ), ( 0 : 19 ), ( -11 : 8 ), 2) )) END

Referências

[Appice et al., 2006] Appice, A., D'Amato, C., Esposito, F. and Malerba, D.: Classification of symbolic objects: A lazy learning approach. Journal of Intelligent Data Analysis vol.10 (2006) pp.301- 324 IOS Press.

[Bacelar-Nicolau,1985] Bacelar-Nicolau, H. The Affinity Coefficient in Cluster Analysis, Methods of Operation Research, v. 53, p. 507-512, Martin J. Bekman et al. (ed), Verlag Anton Hain, Munchen, 1985.

[Bezerra & De Carvalho, 2004]

Bezerra, B. L. D., De Carvalho, F. A. T. A symbolic approach for content-based information filtering. Information Processing Letters, Amsterdam (Holland), v. 92, n.1, p.45-52, 2004.

[Billard & Diday, 2000]

L. Billard and E. Diday. Regression analysis for interval- valued data. In H. A. L. Kiers et al, editor, Data Analysis, Classification and Related Methods, pages 369-374, Berlin, 2000.

[Breiman el al., 1984] Leo Breiman, Jerome H. Friedman, Richard A. Olshen, and Charles J. Stone. Classification and Regression Trees. Wadsworth, 1984.

[Bock, 2000] Hans Hermann Bock. The classical data situation. In Hans- Herman Bock and Edwin Diday, editors, Analysis of Symbolic Data: Exloratory Methods for Extracting Statistical Information from Complex Data, pages 24-38, Germany, 2000. Springer.

[Bock & Diday, 2000] Bock, H.H., Diday, E. Analysis of Symbolic Data. Exploratory Methods for Extracting Statistical Information from Complex Data, series: Studies in Classification, Data Analysis, and Knowledge Organization, v. 15, Springer- Verlag, Berlin, 2000.

[Braga et al., 2000] A. P. Braga, T. B. Ludermir, and A. de Carvalho. Redes Neurais Artificiais - Teoria e Aplicações. LTC, Rio de Janeiro, 2000.

[Ciampi, 1992] A. Ciampi. Constructing prediction trees from data: the recpam approach. Proceedings from the Prague University Summer School on Computacional Aspect of Model Choice, pages 105-152. Verlag, 1992.

Referências Bibliográficas

[Ciampi et al., 2000] A. Ciampi, E. Diday, J. Lebbe, E. Périnel, and R. Vignes. Growing a tree classifier with imprecise data. Pattern Recognition Letters, 21(9):787-803, 2000.

[Cover & Hart, 1967] Cover, T. M. and Hart, P. E.: Nearest neighbor patter lassification, IEEE Trans Inform Theory, 13, 1967, 21-27 [Csiszàr, 1967] Csiszàr, I: Information-type measures of difference of

probability distributions and indirect observations. Studia Scient. MAth. Hung, 2, (1967), 299-318.

[D’Oliveira et al., 2004]

D'Oliveira, S., De Carvalho, F.A.T. and Souza, R.M.C.R.: Classifcation of sar images through a convex hull region oriented approach. In: N. R. Palet al. (Eds.). 11th International Conference on Neural Information Processing (ICONIP-2004), Lectures Notes in Computer Science - LNCS 3316, Springer, (2004), 769-774

[De Carvalho, 1992] De Carvalho, F.A.T.: Méthodes Descriptives en Analyse des Données Symboliques. PhD thesis, Université Paris IX- Dauphine, 1992.

[De Carvalho, 2006] De Carvalho, F.A.T.: Fuzzy c-means clustering methods for symbolic interval data. Pattern recognition Letters, 28, 423- 437, 2006.

[De Carvalho et al., 1999]

De Carvalho, F. A. T., Verde, R., Lechevallier, Y. A dynamic clustering of symbolic objcts based on a context dependent proximity measure. In: IX International Symposium on Applied Stochastic Models and Data analysis. Lisboa: University of Lisboa, p. 237 – 242, 1999.

[De Carvalho et al., 2004]

De Carvalho, F. A. T., Lechevallier, Y. , Souza, R. M. C. R. . A dynamic cluster algorithm based on adaptive Lr distances for quantitative data. 9th Conference of the International Federation of Classification Societies (IFCS2004). New York (USA): Springer-Verlag, 2004. p. 33-42.

[De Carvalho & Diday, 1998]

De Carvalho, F.A.T. and Diday, E.: Indices de proximité entre objects symboliques qui tient compte des contraintes dans l´espace de description. Induction symbolic et numerique à partir de données, Toulouse, 1998. CEPADUES.

[Djouadi & Bouktache, 1997]

Djouadi, A. and Bouktache, E.: Afast algorithm for the nearest-neighbor classier, IEEE Trans. Pattern Anal. Mach. Intell. 19, (3), 1997, 277-282

Referências Bibliográficas

[Duda et al., 2001]. Duda, R. O., Hart, P. E. and Stork, D. G.: Pattern Classificarion, Second Ed., Wiley, New York, 2001

[Esposito et al., 2000] Esposito, F., Malerba, D., Tamma, V. Dissimilarity Measures for Symbolic Objects. In: Bock, H.H., Diday, E. (eds.): Analysis of Symbolic Data. Exploratory Methods for extracting Statistical Information from Complex Data, Series: Studies in Classification, Data Analysis, and Knowledge Organization, Springer-Verlag, Berlin, v. 15, p. 165-185, 2000.

[Fayyad et al., 1996] U. Fayyad, G. Platetsky-Shapiro, and P. Smyth. From data minig to knowledge discovery: an overview. In Advances in Knowledge Discovery and Data Mining, pages 1-34, 1996. [Ferri et al., 1999] Ferri, F. J., Albert, J. V. and Vidal, E.: Considerations about

sample-size sensitivity of a family of edited nearest-neighbor rules, IEEE Trans. Systems Man Cybernet. Part B: Cybernet. 29 (4), 1999, 667-672

[Garden, 1998] S. R. Garden. Building the data warehouse. Communications of the ACM, 41(9):52-60, 1998.

[Gates, 1972] Gates, G. W.: The reduced nearest neighbor rule, IEEE Trans. Inform. Theory, 18, 1972, 431-433

[Gora et al., 2002] Gora, G., Wojna, A.: RIONA: A Classifier Combining Rule Induction and k-NN Method with Automated Selection of Optimal Neighbourhood. Proceedings of the Thirteenth European Conference on Machine Learning, Springer- Verlag, 2430, (2002), 111-123.

[Grother et al., 1997] Grother, P. J., Candela, G. T. and Blue, J. L.: Fast implementarions od nearest neighbor classiers, Pattern Recognition 30, (3), 1997, 459-465

[Guru et al., 2004] Guru, D.S., Kiaranagi, B.B. and Nagabhushan, P.: Multivalued type proximity measure and concept of mutual similarity value useful for clustering symbolic patterns. Pattern recognition Letters, 25, 1203-1213, 2004.

[Hart, 1968] Hart, P. E.: The condensed nearest neighbor rule, IEEE Trans. Inform. Theory, 14, 1968, 515-516

[Ichino & Yaguchi, 1994]

Ichino, M., Yaguchi, H. Generalized minkowski metrics for mixed feature-type data analysis. IEEE Transactions on Systems, Man, and Cybernetics, v. 24, n. 4, p. 698-708, 1994.

Referências Bibliográficas [Johnson & Wichern,

2001] Richard Arnold Johnson and Dean W. Wichern. Applied Multivariate Statistical Analysis. Prentice Hall, fifth edition, 2001.

[Kohonen, 1989] Kohonen, T.: Self-organizarion and Associative Memory, Third Ed., Springer, Heidelberg, Germany, 1989

[Lewis, 2000] Roger J. Lewis. An introduction to classification and regression tree (cart) analysis. Annual Meeting of the Society for Academic Emergency Medicine,

San Francisco, California, 2000.

[Palumbo et al., 2000] F. Palumbo N. Carlo Lauro, R. Verde. Factorial discriminant analysis on symbolic objects. In Hans-Herman Bock and Edwin Diday, editors, Analysis of Symbolic Data: Exploratory Methods for Extracting Statistical Information from Complex Data, pages 212-233, Germany, 2000. Springer.

[Rasson & Lissoir, 2000]

Jean-Paul Rasson and Sandrine Lissoir. Classical methods of discrimination. In Hans-Herman Bock and Edwin Diday, editors, Analysis of Symbolic Data: Exloratory Methods for Extracting Statistical Information from Complex Data, pages 234-240, Germany, 2000. Springer.

[Rossi & Conan-Guez, 2002]

Fabrice Rossi and Brieuc Conan-Guez. Multi-layer perceptron on interval data. Classification, Clustering, and Data Analysis (IFCS 2002), pages 427-434, Cracow, Poland, 2002.

[Rumelhart & McClelland, 1986]

J.L. Rumelhart, D.E.; McClelland. Parallel Distributed Processing: Explorations in the Microstruture of Cognition, volume 1. Cambridge, Mass, 1986.

[Silva et al., 2006] SILVA, Fabio C.d. ; DE CARVALHO, F. T. ; SOUZA, R. M. C. R. ; SILVA, J. Q. . A Modal Symbolic Classifier for Interval Data. In: 13th International Conference on Neural Information Processing - ICONIP2006. Heidelberg (Germany) : Springer, 2006. v. 4233. p. 50-59.

[Simoff, 1996] S. J. Simoff. Handling uncertainty in neural networks: An interval approach. Int. Conf. on Neural Networks, pages 606-610, Washington, 1996. IEEE.

Silva, Fábio César Donato

Classificação supervisionada usando dados simbólicos de semântica modal / Fábio César Donato

Silva.

–

x, 84 folhas : il., fig., tab.

Dissertação (mestrado) – Universidade Federal de Pernambuco. CIn. Ciência da Computação, 2007. Inclui bibliografia e apêndice.

1. Inteligência artificial. 2. Inteligência

computacional. Título.