Bulanık Hedef Programlama Tabanlı Üç Gruplu Sınıflandırma Problemi Yaklaşımı

Bu çalışmada, üç gruplu sınıflandırma probleminin çözümü için bulanık mantık ve matematiksel programlamaya dayalı yeni bir model önerilmiştir. Sınıflandırma problemlerinde ayırma eksenine karşılık gelen kesme değerinin belirlenmesi önem arz etmektedir. Kesme değerinin; asimetrik üçgen bulanık sayı, yamuk bulanık sayı ve gauss bulanık sayı olması durumları incelenmiştir. Önerilen yaklaşım, literatürde sıkça kullanılan 3 gruplu veri setleri kullanılarak Fisher’in Doğrusal Diskriminant Fonksiyonu ve bazı matematiksel programlama yöntemleri ile karşılaştırıldığında daha iyi performans göstermiştir.

Three group classification problem approach based on fuzzy goal programming

In this study, a new fuzzy logic and mathematical programming based model was proposed to solve three-group classificationproblem. Determination of cut-off value, which corresponds to discrimination axis in classification problems, has importance.Status of the cut-off value such as asymmetric triangle fuzzy number, trapezoid fuzzy number and gauss fuzzy number wasexamined. The proposed approach displayed better performance when compared to Fisher's Linear Discriminant Function andsome mathematical programming-based models by using three group data sets used frequently in the literature.

PDF

___

[1] Fisher, R.A., “The use of multiple measurements in taxonomy problems”, Annals of Eugenics, 7: 179-188, (1936).
[2] Smith, C. A., “Some examples of discrimination”, Annals of Eugenics, 13(1): 272-282, (1946).
[3] Freed, N. & Glover, N.,”A linear programming approach to the discriminant problem”, Decision Sciences, 12: 68-74, (1981).
[4] Stam, A. & Ragsdale, C.T., “On the classification gap in mathematical programming-based approaches to the discriminant problem”, Naval Research Logistics, 39: 545- 559, (1992).
[5] Rosen, J.B.,” Pattern separation by convex programming”, Journal of Mathematical Analysis and Applications, 10: 123-134, (1965).
[6] Mangasarian O., “Linear and Nonlinear Separation of patterns by Linear Programming”, Operations Research, 13: 444-452, (1965).
[7] Smith, F.W. “Pattern classifier design by linear programming”, IEEE Transactions on Computers, 17(4): 367-372, (1968).
[8] Grinold, R.C.,” Mathematical programming methods for pattern classification”, Management Sciences, 19: 272- 289,(1972).
[9] Bajgier, S. M.& Hill, A. V.,” An experimental comparison of statistical and linear programming approaches to the discriminant problem”, Decision Sciences, 13: 604–618, (1982).
[10] Lam, K.F., Moy, J.W., “An experimental comparison of some recently developed linear programming approaches to the discriminant analysis”, Computers and Operations Research, 24(7): 593-599, (1997).
[11] Glen, J. J.,” Integer programming methods for normalisation and variable selection in mathematical programming discriminant analysis models”, Journal of Operational Research Society, 50: 1043–1053, (1999).
[12] Lam, K.F., Choo, E.U., Moy, J.W.,” Minimizing deviations from the group mean: A new linear programming approach for the two-group classification problem”, European Journal of Operational Research, 88: 358-367, (1996).