"2012-01-01+01:00"^^ . "Slezsk\u00E1 univerzita v Opav\u011B. Obchodn\u011B podnikatelsk\u00E1 fakulta v Karvin\u00E9" . . . . . . "Karvin\u00E1" . . "Karvin\u00E1" . "Comparison of Different Non-statistical Classification Methods" . "Popelka, Ond\u0159ej" . . . . "In this article, we aim to compare different methods usable for solving classification problems. A substantial number of methods that are not based on mathematical statistics may be used. Exploring these methods is interesting, because they are often capable of solving problems, which are not easily solvable using classificators based purely on mathematical statistics. There are many approaches available such as support vector machines, neural networks, evolutionary algorithms, parallel coordinates, etc. In this article, we concentrate on describing different neural network approaches, parallel coordinates and genetic algorithms. Neural networks come in many flavors (e.g. multi-layer perceptron, non-linear autoregressive networks) and they have achieved some recognition. Genetic algorithms also have been used for classification many times before, but with mixed results. In this article, we describe and evaluate different capabilities of these methods when used for economic data. This for example includes identification of hidden data structures, dealing with outliers and noise." . . "5"^^ . . . "Comparison of Different Non-statistical Classification Methods"@en . . "Hodinka, Michal" . "H\u0159eb\u00ED\u010Dek, Ji\u0159\u00ED" . "127903" . "RIV/62156489:43110/12:00190768!RIV13-GA0-43110___" . "In this article, we aim to compare different methods usable for solving classification problems. A substantial number of methods that are not based on mathematical statistics may be used. Exploring these methods is interesting, because they are often capable of solving problems, which are not easily solvable using classificators based purely on mathematical statistics. There are many approaches available such as support vector machines, neural networks, evolutionary algorithms, parallel coordinates, etc. In this article, we concentrate on describing different neural network approaches, parallel coordinates and genetic algorithms. Neural networks come in many flavors (e.g. multi-layer perceptron, non-linear autoregressive networks) and they have achieved some recognition. Genetic algorithms also have been used for classification many times before, but with mixed results. In this article, we describe and evaluate different capabilities of these methods when used for economic data. This for example includes identification of hidden data structures, dealing with outliers and noise."@en . . . "Trenz, Old\u0159ich" . "5"^^ . . "P(GAP403/11/2085)" . "Comparison of Different Non-statistical Classification Methods"@en . . "RIV/62156489:43110/12:00190768" . "Comparison of Different Non-statistical Classification Methods" . "corporate performance; parallel coordinates; decision trees; sustainability reporting; neural networks; classification"@en . . "6"^^ . "Proceedings of the 30th International Conference Mathematical Methods in Economics 2012" . "\u0160tencl, Michael" . . . "978-80-7248-779-0" . . "43110" . "[B1AC0966BC1C]" . . . .