Probabilities in the Problem of Classification - Probabilistic Composition of Preferences, Theory and Applications - page 93

Table 8.8 PseudoF and
weighted PseudoF values
for cars classi cation

Statistics Triangular

classi cation

Normal
classi cation

PseudoF

2.22 1.84

WPseudoF

3.19 2.44

to Class 3. The weights for the computation of PseudoF are those employed to
create the classi

cations.
Table 8.8 shows that, with a relatively larger between classes variance, the
statistics present better values for the first classi cation.

References

Almeida-Dias, J., Figueira, J. R., & Roy, B. (2010). ELECTRE TRI-C: a multiple criteria sorting
method based on characteristic reference actions. European Journal of Operational Research,
204(3), 565 - 580.
Almeida-Dias, J., Figueira, J. R., & Roy, B. (2012). A multiple criteria sorting method where each
category is characterized by several reference actions: the ELECTRE TRI-nC method.
European Journal of Operational Research, 217(3), 567 - 579.
Calinski, T., & Harabasz, J. (1974). A Dendrite method for cluster analysis. Communications in
Statistics, 3,1 27.

Roy, B. (1968) Classement et choix en presence de points de vue multiples (la method
ELECTRE), pp. 57

-

- 75, La Revue d ' Informatique et de Recherche Op é rationelle (RIRO).

Sant ' Anna, A. P., Costa, H. G., & Pereira, V. (2012). CPP-TRI: um m é todo de classica çã o

ordenada baseado em composi çã o probabil í stica. Relat ó rios de Pesquisa em Engenharia de

117.
Yu, W. (1992). ELECTRE TRI — Aspects methodologiques et guide d ' utilisation. Document du
LAMSADE: Universit

Produ çã o, 12(8), 104 -

de Paris-Dauphine, Paris.
Zadeh, L. (1965). Fuzzy Set. Information Science, 8(3), 338

é

- 353.