• Nenhum resultado encontrado

COMPSTAT2008 AM ASF MC

N/A
N/A
Protected

Academic year: 2018

Share "COMPSTAT2008 AM ASF MC"

Copied!
1
0
0

Texto

(1)

Combining Models in Discrete Discriminant

Analysis in the Multiclass Case

Anabela Marques1

Ana Sousa Ferreira2

and Margarida Cardoso3

1

Barreiro College of Tecnology, Set´ubal Polytechnic, Portugal CEAUL, Project of Multivariate Data Analysis and Modelling

anabela.marques@estbarreiro.ips.pt

2

LEAD, FPCE, University of Lisbon, Lisbon, Portugal

CEAUL, Project of Multivariate Data Analysis and Modelling

asferreira@fpce.ul.pt

3

Dep. of Quantitative Methods, ISCTE Business School, Lisbon, Portugal, UNIDE,margarida.cardoso@iscte.pt

Abstract. The idea of combining models in Discrete Discriminant Analysis (DDA) is present in a growing number of papers which aim to obtain more robust and more stable models than any of the competing ones. This seems to be a promising ap-proach since it is known that different DDA models perform differently on different subjects (Brito et al.(2006)). In particular, this will be a more relevant issue if the groups are not well separated, which often occurs in practice.

In the present work a new methodological approach is suggested which is based on DDA models’ combination. The multiclass problem is decomposed into several dichotomous problems that are nested in a hierarchical binary tree (Sousa Ferreira (2000), Brito et al. (2006)) and at each level of the binary tree a new combining model is proposed to derive the decision rule. This combining model is based on two well known models in the literature - the First-order Independence Model (FOIM) and the Dependence Trees Model (DTM) (Celeux and Nakache (1994)).

The MATLAB software is used for the algorithms’ implementation and the pro-posed approach is illustrated in a DDA application.

Keywords: Combining model, Discrete Discriminant Analysis , First-order Independence Model, Dependence Trees Model

References

BRITO, I., CELEUX, C. and SOUSA FERREIRA, A. (2006): Combining methods in supervised classification: A comparative study on discrete and continuous problems.REVSTAT - Statistical Journal, Vol. 4(3), 201-225.

CELEUX, G. and NAKACHE, J. P. (1994): Analyse Discriminante sur Variables Qualitatives. G. Celeux et J. P. Nakache ´Editeurs, POLYTECHNICA. SOUSA FERREIRA, A. (2000):Combina¸c˜ao de Modelos em An´alise Discriminante

Referências

Documentos relacionados

Também os pais dos alunos de ambos os colégios adotam em casa esta ética de vida, transmitindo aos filhos os valores do mérito, do esforço e do trabalho e sensibilizan- do-os para

Conclusions: Single dose intravesical hyluronic acid instillation reduces inflammatory cell infiltration and the severity of bladder inflammation in the rat model of bladder

Tidal analysis of 29-days time series of elevations and currents for each grid point generated corange and cophase lines as well as the correspondent axes of the current ellipses

Posteriormente, aplicamos o método de minimax e a teoria de Morse em dimensão infinita para demonstrar que o problema não homogêneo possui pelo menos três soluções não triviais..

Based on the appreciation of the deinitions of powerlessness and the critical attributes essential to this concept evidenced in the literature is constituted from the

Resumo Relatório de Estágio no âmbito do Mestrado em Ensino de Artes Visuais do 3.º Ciclo no Ensino Básico e do Ensino Secundário, iniciado em Setembro de 2010, na Escola

The models proposed include a constitutive model for the material being cut based on the work of Weber, a model for the shear plane based on Merchants model,

Figure 3.1 illustrates a Model Altering Model Morphism between two models (the source “A” and target “B” models), where when it is applied to the source model it results