2020 | 30 | 1 | 119-130
Article title

On the binary classification problem in discriminant analysis using linear programming methods

Title variants
Languages of publication
This paper is centred on a binary classification problem in which it is desired to assign a new object with multivariate features to one of two distinct populations as based on historical sets of samples from two populations. A linear discriminant analysis framework has been proposed, called the minimised sum of deviations by proportion (MSDP) to model the binary classification problem. In the MSDP formulation, the sum of the proportion of exterior deviations is minimised subject to the group separation constraints, the normalisation constraint, the upper bound constraints on proportions of exterior deviations and the sign unrestriction vis-à-vis the non-negativity constraints. The two-phase method in linear programming is adopted as a solution technique to generate the discriminant function. The decision rule on group-membership prediction is constructed using the apparent error rate. The performance of the MSDP has been compared with some existing linear discriminant models using a previously published dataset on road casualties. The MSDP model was more promising and well suited for the imbalanced dataset on road casualties.
Physical description
  • Nnamdi Azikiwe University, PMB 5025 Awka, Nigeria
  • Nnamdi Azikiwe University, PMB 5025 Awka, Nigeria
  • ADEGBOYE O.S., The optimal classification rule for exponential populations, Austr. J. Stat., 1993, 35 (2), 185–194.
  • ABRAMOVICH F., PENSKY M., Classification with many classes: challenges and pluses, J. Mult. Anal., 2019, 174. Available at 10.1016/j.jmva.2019.104536
  • AWOGBEMI C.A., ONYEAGU S.I., Distribution of errors of misclassification for the linear discriminant function (a case of Edgeworth series non-normal distribution), Math. Theory Model., 2018, 8 (4), 30–44.
  • ERENGUC S.S., KOEHLER G.J., Survey of mathematical programming models and experimental results for linear discriminant analysis, Manage. Dec. Econ., 1990, 11 (4), 215–225.
  • FALANGIS K., Mathematical programming models for classification problems with applications to credit scoring, PhD Thesis, The University of Edinburgh, Edinburgh 2013.
  • FALANGIS K., GLEN J.J., Heuristics for feature selection in mathematical programming discriminant analysis models, J. Oper. Res. Soc., 2010, 61, 804–812.
  • FREED N., GLOVER F., Evaluating alternative linear programming models to solve the two-group discriminant problem, Dec. Sci., 1986, 17, 151–162.
  • GAYNANOVA I., WANG T., Sparse quadratic classification rules via linear dimension reduction, J. Multiv. Anal., 2019, 169, 278–299.
  • GLEN J.J., Classification accuracy in discriminant analysis: a mixed integer programming approach, J. Oper. Res. Soc., 2001, 52 (3), 328–339.
  • GLOVER F., Improved linear programming models for discriminant analysis, Dec. Sci., 1990, 21, 771–785.
  • GOCHET W., STAM A., SRINIVASAN V., CHEN S., Multigroup discriminant analysis using linear programming, Oper. Res., 1997, 45 (2), 213–225.
  • KOEHLER G.J., Considerations for mathematical programming models in discriminant analysis, Manage. Dec. Econ., 1990, 11 (4), 227–234.
  • LAM K.F., MOY J.W., Improved linear programming formulations for the multi-group discriminant problem, J. Oper. Res. Soc., 1996, 47 (12), 1526–1529.
  • 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, Eur. J. Oper. Res., 1996, 88, 358–367.
  • LIITTSCHWAGER J.M., WANG C., Integer programming solution of a classification problem, Manage. Sci., 1978, 24 (14), 1515–1525.
  • LIU Y.-H., MALONEY J., Discriminant analysis and linear programming, Int. J. Math. Edu. Sci. Technol., 1997, 28 (2), 207–210.
  • MAKINDE O.S., On misclassification probabilities of linear and quadratic classifiers, Afr. Stat., 2016, 11 (1), 943–953.
  • RENCHER A.C., Method of Multivariate Analysis (2nd Ed.), Wiley, New York 2002.
  • STAM A., JONES D.G., Classification performance of mathematical programming techniques in discriminant analysis. Results for small and medium sample sizes, Manage. Dec. Econ., 1990, 11 (4), 243–253.
  • ZIARI H.A., LEATHAM D.J., ELLINGER P.N., Development of statistical discriminant mathematical programming model via resampling estimation techniques, Am. J. Agr. Econ., 1997, 79 (4), 1352–1362
Document Type
Publication order reference
YADDA identifier
JavaScript is turned off in your web browser. Turn it on to take full advantage of this site, then refresh the page.