The Sensitivity of machine learning techniques to variations in sample sizea comparative analysis

  1. Andrés Suárez, Javier de 1
  2. Lorca Fernández, Pedro 1
  3. Fernández Combarro Álvarez, Elías 1
  1. 1 Universidad de Oviedo
    info

    Universidad de Oviedo

    Oviedo, España

    ROR https://ror.org/006gksa02

Revista:
The International Journal of Digital Accounting Research

ISSN: 1577-8517

Año de publicación: 2002

Volumen: 2

Número: 4

Páginas: 131-155

Tipo: Artículo

DOI: 10.4192/1577-8517-V2_5 DIALNET GOOGLE SCHOLAR lock_openArias Montano editor

Otras publicaciones en: The International Journal of Digital Accounting Research

Resumen

A comparative analysis of the performance of some well-known classification techniques (Discriminant Analysis, Quinlan's See5, and Neural Networks) and certain machine learning systems of recent development (ARNI, FAN and SVM) is conducted. The chosen classification task is the forecasting of the level of efficiency of Spanish commercial and industrial companies. Assignment of the firms is made upon the basis of a set of financial ratios, which make a high dimension feature space with low separability degree. In the present research the effects on the accuracy of variations of each technique in the estimation sample size are measured. The main results suggest that ARNI and See5 yield the best results, even with small sample sizes.

Fuente de los datos: Dialnet