A methodological approach for pattern recognition system using discriminant analysis and artificial neural networks
Abstract:
In this work it is presented a methodology for the development of a pattern recognition system using classification methods as discriminant analysis and artificial neural networks. In this methodology, the statistical analysis is contemplated, with the purpose of retaining the observations and the important characteristics that can produce an appropriate classification, and allows, as well, to detect outliers' observations, multicolinearity between variables, among other things.
Año de publicación:
2005
Keywords:
- classification
- artificial neural networks
- pattern recognition
- Discriminant analysis
Fuente:
scopus
Tipo de documento:
Article
Estado:
Acceso restringido
Áreas de conocimiento:
- Red neuronal artificial
- Ciencias de la computación
- Ciencias de la computación
Áreas temáticas:
- Ciencias de la computación