A Binary Classification Algorithm Based on Polyhedral Conic Functions

Abstract

Data classification is one of the main techniques of data mining. Different mathematical programming approaches of the data classification were presented in recent years. A technique that uses polyhedral conic functions (PCF) is an effective method for data classification. We present a modified classification algorithm based on PCF functions. Results of numerical experiments on real-world and synthetic data sets demonstrate that the proposed approach is efficient for solving binary data classification problems.

Keywords

• Mathematical Programming
• Polyhedral Conic Functions
• Classification
• Clustering

Çokyüzlü Konik Fonksiyonlar Temelli Bir İkili Sınıflandırma Algoritması

Abstract

Veri sınıflandırma, veri madenciliğinin önemli tekniklerinden birisidir. Son yıllarda veri sınıflandırması için farklı matematiksel programlama yaklaşımları sunulmuştur. Çokyüzlü konik fonksiyonları kullanan bir teknik veri sınıflandırması için efektif bir yöntem olmuştur. Bu çalışmada çokyüzlü konik fonksiyonları temel alan geliştirilmiş bir sınıflandırma algoritması sunulmuştur. Gerçek hayat ve sentetik veri kümeleri üzerinde yapılan sayısal deney sonuçları göstermektedir ki sunulan yaklaşım ikili veri sınıflandırma problemlerinin çözümünde etkili olmuştur.

Keywords

• Matematiksel Programlama
• Çokyüzlü Konik Fonksiyonlar
• Sınıflandırma
• Kümeleme

References

1. Anderberg M.R., “Cluster Analysis for Applications”, Academic Press, New York, (1973).
2. Astorino A., Gaudioso M., “Polyhedral Separability through Successive LP”, Journal of Optimization Theory and Applications, Vol:112, No:2, February, (2002), pp. 265-293.
3. Bagirov A.M., ” Max Min Separability”, Optimization Methods and Software, Vol:20, No:2-3, April-June, (2005), pp. 277-296.
4. Bagirov A.M., Mardaneh K., “Modified global k-means algorithm for clustering in gene expression data sets”, WISB '06 Proceedings of the 2006 workshop on Intelligent systems for bioinformatics – Vol: 73 , (2006), pp. 23-28.
5. Bagirov A.M., Ugon J., “Supervised Data Classification via Max-Min Separability”, Continous Optimization,Applied Optimization, Vol:99, (2005), pp. 175-207.
6. Bagirov A.M., Ugon J., Webb D., Karasözen B., “ Classification through incremental max–min separability”, Pattern Analysis and Applications, Vol:14, Issue: 2, (2011), pp.16518-174.
7. Bagirov A.M., Ugon J., Webb D., Öztürk G., Kasımbeyli R, “A novel piecewise linear classifier based on polyhedral conic and max-min separabilities, TOP, (2011) DOI: 10.1007/s11750-011-0241-5.
8. Bennett K.P., Mangasarian O.L, “Robust linear programming discrimination of two linearly inseparable sets”, Optimization methods and software 1 (1), (1992), pp. 23-34,.

9. Gasimov R.N., Öztürk G., “Separation via polyhedral conic functions”, Optimization Methods and Software, Vol:21,No:4, August, (2006), pp.527-540,.
10. -
11. Kohavi R., “A study of cross-validation and bootstrap for accuracy estimation and model selection”, (1995), International Joint Conference on Artificial Intelligence.
12. Kusiak A., “Data Analysis: Models and Algorithms”, Proc. SPIE Vol. 4191, (2001), pp. 1-9,.
13. Liittschwager J. M. and Wang C., Management Science Vol. 24, No. 14 Oct., (1978), pp. 1515-1525.
14. Mangasarian O.L., “Linear and nonlinear separation of patterns by linear programming”, Operations Research, Volume 13, Issue 3 (May-June), (1965), pp. 444-452,.
15. Raginsky M., “Binary Classification”, 2011/3/14.
16. Rosen J.B., “ Pattern separation by convex programming”, Stanford Univ. Calif. Applied Mathematics and Statistics Labs, (1963).
17. Vapnik V.,” The nature of statistical learning theory”, (1995), Springer Verlang, New York.