Abstract
In this paper, problems of mathematical diagnostics are considered. The most popular approach to these problems is based on statistical methods. In this paper, the author treats the mentioned problems by means of optimization. This approach can be useful in the case where statistical characteristics of the database are unknown or the database is not sufficiently large. In this paper, a nonsmooth model is used where it is required to separate two sets, whose convex hulls may intersect. A linear classifier is used to identify the points of two sets. The quality of identification is evaluated by the so-called natural functional, based on the number of misclassified points. It is required to find the optimal hyperplane, which minimizes the number of misclassified points by means of the translation and rotation operations. Since the natural functional (number of misclassified points) is discontinuous, it is suggested to approximate it by some surrogate functional possessing at least the continuity property. In this paper, two surrogate functionals are introduced and studied. It is shown that one of them is subdifferentiable, and the second one is continuously differentiable. It is also demonstrated that the theory of exact penalization can be employed to reduce the given constrained optimization problems to an unconstrained one. Numerical methods are constructed, where the steepest descent directions of the surrogate functionals are used to minimize the natural one. Necessary conditions for a minimum are formulated for both surrogate functionals.
Similar content being viewed by others
References
Michie, D., Spiegelhalter, D.J., Taylor, C.C.: Machine Learning: Neural and Statistical Classification. Ellis Horwood, Chichester (1994)
Quinlan, J.R.: Programs for Machine Learning. Morgan Kaufmann, San Mateo (1993)
Astorino, A., Fuduli, A.: Support vector machine polyhedral separability in semisupervised learning. J. Opt. Theory Appl. 164(3), 1039–1050 (2015)
Bradley, P.S., Mangasarian, O.L.: Feature selection via concave minimization and support vector machines, machine learning. In: Proceedings of the Fifteenth International Conference (ICML’98), Morgan Kaufmann, San Francisco, California, pp. 82–90 (1998)
Bradley, P.S., Mangasarian, O.L.: Massive data discrimination via linear support vector machines. Opt. Methods Softw. 13, 1–10 (2000)
De Coste, D., Scholkopf, B.: Training invariant support vector machines. Mach. Learn. 46, 161–190 (2002)
Lee, Y.-J., Mangasarian, O.L.: SSVM: a smooth support vector machine for classification. Comput. Optim. Appl. 20(1), 5–22 (2001)
Mangasarian, O.L.: Linear and nonlinear separation of patterns by linear programming. Oper. Res. 13, 444–452 (1965)
Mangasarian, O.L.: Misclassification minimization. J. Glob. Optim. 5, 309–323 (1994)
Mangasarian, O.L.: Mathematical programming in data mining. Data Min Knowl Discov 1, 183–201 (1997)
Vapnik, V.: The Nature of Statistical Learning Theory, 334 p. Springer, New York (2000)
Anderberg, M.R.: Cluster Analysis for Applications. Academic Press, New York (1973)
Bhuyan, N.J., Raghavan, V.V., Venkatesh, K.E.: Genetic algorithms for clustering with an ordered representation. In: Proceedings of the Fourth International Conference on Genetic Algorithms, pp. 408–415 (1991)
Hansen, P., Jaumard, B.: Cluster analysis and mathematical programming. Math. Program. 79, 191–215 (1997)
Jain, A.K., Murty, M.N., Flynn, P.J.: Data clustering: a review. ACM Comput. Surv. 31, 264–323 (1999)
Mirkin, B.: Mathematical Classification and Clustering, 448 p. Kluwer Academic Publishers, Dordrecht (1996)
Fisher, R.A.: Contributions to Mathematical Statistics. Wiley, New York (1950)
Nagy, G.: State of the art in pattern recognition. Proce. IEEE 56, 836–862 (1968)
Cristianini, N., Shawe-Taylor, J.: An Introduction to Support Vector Machines and Other Kernel Based Methods, 168 p. Cambridge University Press, Cambridge (2000)
Scholkopf, B., Smola, A.: Learning with Kernels. The MIT Press, Cambridge (2002)
Bagirov, A.M., Rubinov, A.M., Yearwood, J.: A global optimization approach to classification. Optim. Eng. 3, 129–155 (2002)
Bagirov, A.M., Rubinov, A.M., Yearwood, J.: A heuristic algorithm for feature selection based on optimization techniques. In: Abbas, H., Newton, C., Sarker, R. (eds.) Heuristic and Optimization for Knowledge Discovery, pp. 13–26. Idea Publishing Group, Hershey (2002)
Bennett, K.P., Mangasarian, O.L.: Robust linear programming discrimination of two linearly inseparable sets. Optim. Method Softw. 1, 23–34 (1992)
Highleyman, W.H.: Linear decision functions with applications to pattern recognition. Proc. IRE 50, 1501–1514 (1962)
Demyanov, V.F.: Identification of points of two convex sets. Vestnik S.-Petersburg Univ. Ser. I 3(17), 14–20 (2001)
Demyanov, V.F.: Extremum Conditions and Variational Problems, 136 p. SPbSU, St. Petersburg (In Russian) (2000)
Demyanov, V.F., Rubinov, A.M.: Foundation of Nonsmooth Analyses and Quasidifferential Calculus, 431 p. Nauka, Moscow (In Russian) (1990)
Demyanov, V.F., Vasiliev, L.V.: Nondifferential Optimization. Springer, New York (1986)
Karmanov, V.G.: Mathematical Programming. Mir Publishers, Moscow (1989)
Polyak, B.T.: Introduction to Optimization, 384 p. Nauka, Moscow (In Russian) (1983)
Pshenichny, B.N., Danilin, Y.M.: Numerical Methods in Extremal Problems, 2nd edn, 320 p. Mir Publishers, Moscow (2009)
Vasil’ev, F.P.: Numerical Method for Solving Extremal Problems, 550 p. Nauka, Moscow (In Russian) (1988)
Anan’ev, K.I., Demyanova, V.V., Dem’yanov, V.F., Kokorina, A.V., Svistun, S.Ya., Stegalin, I.S.: Optimization methods in diagnostics problems. Vestnik S.-Petersburg Univ. Ser. Prikl. Mat. Inform. Prots. Upr. 10(3), 312 (2011)
Bagirov, A.M., Rubinov, A.M., Soukhoroukova, N.V., Yearwood, J.: Unsupervised and supervised data classification via nonsmooth and global optimization. Top, Sociedad de Estadistica e Investigacion Operativa, Madrid, Spain 11(1), 1–75 (2003)
Demyanov, V.F.: Mathematical diagnostics via nonsmooth analysis. OMS 20, 197–218 (2005)
Demyanov, V.F., Astorino, A., Gaudioso, M.: Nonsmooth problems in mathematical diagnostics. In: Hadjisavvas, N., Pardalos, P.M. (eds.) Advances in Convex Analysis and Global Optimization; Nonconvex Optimization and Its Applications, vol. 54, pp. 11–30. Kluwer Academic Publishers, Dordrecht (2001)
Demyanova, V.V., Demyanov, V.F.: One-dimensional identification problem and ranking parameters. J. Glob. Optim. 48, 29–40 (2010)
Demyanova, V.V., Demyanov, V.F.: The method of virtual experts in mathematical diagnostics. J. Glob. Optim. 35(2), 215–234 (2006)
Grigor’eva, X. V.: Surrogate functionals in diagnosis problems. Vestnik S.-Petersburg Univ. Ser. 10. Prikl. Mat. Inform. Prots. Upr., Issue 1, 34–43 (2009)
Kokorina, A. V.: Unsupervised and supervised Data Classification Via Nonsmooth and Global Optimization. Top, Volume 11, Number 1, Sociedad de Estadistica e Tnvestigacion Operativa, Madrid, Spain, pp. 86–89 (2003)
Malozemov, V.N., Cherneutsanu, E.K.: The best linear separation of two sets. Constructive nonsmooth analysis and related topics. Spring. Optim. Appl. 87, 175–183 (2014)
Bache, K., Lichman, M.: UCI Machine Learning Repository. University of California, School of Information and Computer Science, Irvine, CA (2013). http://archive.ics.uci.edu/ml
Acknowledgments
I would like to thank Dr. N. Sukhorukova (Australia, Melbourne, Swinburne University of Technology) for useful comments.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Grigor’eva, X.V. Approximate Functions in a Problem of Sets Separation. J Optim Theory Appl 171, 550–572 (2016). https://doi.org/10.1007/s10957-015-0766-0
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10957-015-0766-0