Neural modelling of ranking data with an application to stated preference data

Catherine Krier, Michel Mouchart, Abderrahim Oulhaj

Abstract


Although neural networks are commonly encountered to solve classification problems, ranking data present specificities which require adapting the model. Based on a latent utility function defined on the characteristics of the objects to be ranked, the approach suggested in this paper leads to a perceptron-based algorithm for a highly non linear model. Data on stated preferences obtained through a survey by face-to-face interviews, in the field of freight transport, are used to illustrate the method. Numerical difficulties are pinpointed and a Pocket type algorithm is shown to provide an efficient heuristic to minimize the discrete error criterion. A substantial merit of this approach is to provide a workable estimation of contextually interpretable parameters along with a statistical evaluation of the goodness of fit.

Full Text:

PDF (English)

References


B. APOLLONI, I. ZOPPIS, R. ALEANDRI, A. GALLI, (2001), A neural network based procedure for forecasting bovine semen motility, “Statistica”, LXI (2), 301-313.

M. BEUTHE, CH. BOUFFIOUX, J. DE MAEYER, G. SANTAMARIA, M. VANDRESSE, E. VANDAELE, F. WITLOX, (2005), A multi-criteria methodology for stated preferences among freight transport alternatives, in

AURA REGGIANI and LAURIE SCHINTLER (ed.) Methods and Models in Transport and Communications: Cross Atlantic Perspectives, Berlin: Springer Verlag.

M. BEUTHE M., CH. BOUFFIOUX, C. KRIER, M. MOUCHART, (2008), A comparison of conjoint, multicriteria, conditional logit and neural network analyses for rank-ordered preference data, chap. 9 in M.

BEN-AKIVA, H. MEERSMAN and E. VAN DE VOORDE (ed.) Recent Developments in Freight Transport Modelling: Lessons from the Freight Sector, Emerald Group Publishing Limited, 157-178.

M. BEUTHE, G. SCANNELLA, (2001), Comparative analysis of UTA multi-criteria methods, ‘European Journal of operational research’, 130 (2): 246-262.

E. BIGANZOLI, P. BORACCHI, I. POLI, (2000), Reti neurali artificiali per lo studio di fenomeni complessi: limiti e vantaggi delle applicazioni in biostatistica, ‘Statistica’, LX (4), 723-734

C.M. BISHOP, (1995), Neural Networks for Pattern Recognition, Oxford: Oxford University Press.

S.I. GALLANT, (1996), Perceptron based learning algorithms, ‘IEEE Transactions on Neural Networks’, 1:179-191.

S. HAYKIN, (1999), Neural Networks: a comprehensive foundation, 2nd ed., Prentice-Hall, Inc.

A. OULHAJ, M. MOUCHART, (2003), Partial sufficiency with connection to the identification problem, ‘Metron’, LXI (2), 267-683.

M. PILLATI, (2001), Le reti neurali in problemi di classificazione: una soluzione basata sui metodi di segmentazione binaria, ‘Statistica’, LXI (3), 407-421 H.R. VARIAN, (1992), Microeconomic Analysis, 3rd ed., Norton & Company, New-York.




DOI: 10.6092/issn.1973-2201/3646