Categorization of Countries with Artificial Neural Networks and Support Vector Machines

Yıl 2023, Cilt: 1 Sayı: 1, 36 - 45, 30.12.2023

Gökhan Korkmaz

Öz

In this study, the possibilities of ranking or classifying countries, which are generally made using panel data analysis, are investigated using artificial intelligence models. For this, countries are classified in terms of unemployment, inflation, GDP Growth Rate, 5-year GDP Growth Rate, Foreign Direct Investment (FDI) Input and Job Freedom. Artificial Neural Networks (ANN), Support Vector Machines (SVM) and statistically Logistic Regression (LR) methods were used for classification. In the analyzes repeated ten times, LR (average 62.4%) gave the best result and SVM (2%) gave the lowest standard deviation.
The results obtained are promising for modern methods, but modern artificial intelligence methods, which have become an alternative to traditional methods in almost every field, are still behind traditional methods in this field. In order for modern methods to be an alternative to traditional methods in this regard, they need to further develop their theories (on matters such as the curse of dimension) or adapt the data structures used on the subject to these methods.

Anahtar Kelimeler

Country Classification, Artificial Neural Network, Support Vector Machines.

Kaynakça

• Becerra-Fernandez, I., Zanakis, S. H., and Walczak, S. (2002). Knowledge discovery techniques for predicting country investment risk. Computers and Industrial Engineering, 43(4), 787–800.
• Cherkassky, V. S., and Mulier, Filip. (2007). Learning from data : concepts, theory, and methods. Published: John Wiley&Sons.
• Cura, T. (2008). Modern sezgisel teknikler ve uygulamaları. İstanbul: Papatya Publisher.
• Huang, Z., Chen, H., Hsu, C. J., Chen, W. H., and Wu, S. (2004). Credit rating analysis with support vector machines and neural networks: a market comparative study. Decision Support Systems, 37(4), 543–558.
• LaValley, P., M. (2008). Logistic Regression. Circulation. 117(18), 2395-2399.
• Lee, T. S., and Chen, I. F. (2005). A two-stage hybrid credit scoring model using artificial neural networks and multivariate adaptive regression splines. Expert Systems with Applications, 28(4), 743–752.
• Öztemel, E. (2003). Yapay sinir ağları. İstanbul: Papatya Publisher.
• Patuelli, R., Reggiani, A., Nijkamp, P., and Blien, U. (2007). New neural network methods for forecasting regional employment. Research. 1(1), 7–30.
• Shen, J. (2005). Fusing support vector machines and soft computing for pattern recognition and regression. https://search.proquest.com/openview/d2b5e228d906f45d2b282e084f88c3ca/1?pq￾origsite=gscholarandcbl=18750anddiss=y
• Şimşek Gürsoy, U. T. (2009). Veri madenciliği ve bilgi keşfi. İstanbul: Pegem Akademi Publisher.
• Theodoridis, S., and Koutroumbas, K. (2009). Pattern recognition. Burlington: Elseiver Inc.
• Vapnik V (2022). I. transactions on neural, and 1999, undefined. (n.d.). An overview of statistical learning theory. Ieeexplore.Ieee.Org. Retrieved December 9, 2022, from https://ieeexplore.ieee.org/abstract/document/788640/.
• Vellido, A., Lisboa, P. J. G., and Vaughan, J. (1999). Neural networks in business: a survey of applications (1992–1998). Expert Systems with Applications, 17(1), 51–70.
• Yim, J., and Mitchell, H. (2005). Comparison of country risk models: hybrid neural networks, logit models, discriminant analysis and cluster techniques. Expert Systems with Applications, 28(1), 137–148.