Hybrid approach to supporting decision making processes in companies

• Autorzy: Pietruszkiewicz W. , Twardochleb M. , Roszkowski M.
• Języki publikacji: EN

Abstrakty

EN

This article presents the advantages of hybrid approach to the support decision making by analyzing three areas of business decision problems, solved by combination of well-known algorithms into the new hybrid constructions: cascade optimization hybrid, parallel classification hybrid and hybrid multicomponent attribute selection. Each of them solved a different problem: the cascade optimization hybrid allowed for finding an extreme of a composite objective function, the parallel classification hybrid was used to choose a proper class through voting, the multicomponent attribute selection robustly chose significant decision variables. A hybrid approach to the problem of supporting the decision making processes is more effective than using each of the component methods alone, even for the sophisticated ones. A combination of several methods with different characteristics and performance makes it possible to take advantages of their strong sides and simultaneously eliminate the weak ones, resulting in a better computational support of decision making.

Słowa kluczowe

EN

data mining; cascade optimization hybrid; parallel classification hybrid; hybrid multicomponent attribute selection

• Wydawca: Systems Research Institute, Polish Academy of Sciences
• Czasopismo: Control and Cybernetics
• Rocznik: 2011
• Tom: Vol. 40, no 1
• Strony: 125--143
• Opis fizyczny: Bibliogr. 15 poz., wykr.

Twórcy

autor

Pietruszkiewicz W.

autor

Twardochleb M.

autor

Roszkowski M.

• West Pomeranian University of Technology, Faculty of Computer Science and Information Technology, Żołnierska 49, 71-711 Szczecin, Poland,

Bibliografia

• BAGIROV, A.M. and RUBINOV, A.M. (2004) Local Optimization with Global Multidimensional Search for Descent. School of Information Technology and Mathematical Science, University of Ballarat.
• BREIMAN, L., FRIEDMAN J., STONE Cn.J. and OLSHEN R.A. (1984) Classification and Regression Trees. Chapman and Hall/CRC.
• CORTEZ, P., CERDEIRA, A., ALMEIDA, F., MATOS, T. and REIS, J. (2009) Modelling wine preferences by data mining from physicochemical properties. Decision Support Systems 47 (4), 547-553.
• DUCH, W., WIECZOREK, T., BIESIADA, J. and BLACHNIK, M. (2004) Comparison of feature ranking methods based on Information entropy. Proceeding of International Joint Conference on Neural Networks. Proc. of International Joint Conference on Neural Networks (IJCNN), Budapest 2004, IEEE Press, 1415-1420.
• DUCH, W., KORBICZ, J., RUTKOWSKI, L. and TADEUSIEWICZ, R. (eds.) (2000) Biocybernetics and biomedical engineering. Neural networks. Warsaw, Exit, vol.6 (in Polish).
• FRIEDMAN, J. (1991) Multivariate adaptive regression splines (with discussion). Annals of Statistics, 19, 1-141.
• GUYON, I. and ELISSEE, A. (2003) An introduction to variable and feature selection. Journal of Machine Learning Research 3, 1157-1182.
• HALL, M.A. and SMITH, L.A. (1999) Feature Selection for Machine Learning: Comparing a Correlation-based Filter Approach to the Wrapper. Proceedings of the Twelfth International FLAIRS Conference. AAAI Press, 235-239.
• PIETRUSZKIEWICZ, W. (2010) The Hybrid Feature Selection for the Prediction of Household Bankruptcy. Proceeding of 5th International Conference on Neural Networks and Artificial Intelligence. Brest’ State Technical University, Brest.
• PINTER, J.D. (2002) Global and Convex Optimization in Modelling Environments: Compiler-Based, Excel, and Mathematica Implementations. Research Report, PCS Inc. www.optimization-online.org/DB_FILE/2002/08/523.pdf.
• ROSZKOWSKI, M. (2006) A method for identification of objects state with the use of neural networks based on the example of classifying the e-market clients. Doctoral dissertation, Faculty of Information Technology, Technical University of Szczecin, Szczecin.
• ROZENBERG, L. and PIETRUSZKIEWICZ, W. (2008) The methodology of diagnosis and prognosis of household bankruptcy (in Polish). Difin, Szczecin.
• TWARDOCHLEB, M.R. and RYCHCICKI, R. (2009) Efficiency of hybrid optimization method in solving tasks of verified characteristics (in Polish). Metody Informatyki Stosowanej, 4, 135-145.
• WITKOWSKA, D. (2002) Artificial neural networks and statistical methods. Selected financial problems. C.H. Beck, Warsaw (in Polish).
• ZHENG, Z., WU, X. and SRIHARI, R. (2004) Feature selection for text categorization on imbalanced data. ACM SIGKDD Explorations Newsletter, 6 (1). ACM, New York, 80-89.