Local Representation Neural Networks for Feature Selection
M. Mar Abad Grau* and L. Daniel Hernandez Molinero**
*Departamento de Lenguajes y Sistemas Informaticos Universidad de Granada, Espana
**Departamento de Informatica y Sistemas, Universidad de Murcia, Espana
Received:June 25, 1998Accepted:May 19, 1999Published:August 20, 1999
Keywords:Neural network, Superrised learning, Feature selection
Pruning methods for feature selection in neural networks start out from the idea that the representation of the data must evolve from a distributed representation of the information to a more localised representation which will represent the skeleton of the network, needing long training times imposed by the back propagation algorithm. Even the quasi-Newton algorithm spent a long computation time. We propose a three-layer network based on local representation with a step-threshold function and an algorithm called Direct Method for Structural Learning, both allow a very fast pruning of superfluous attributes.
Cite this article as:M. Grau and L. Molinero, “Local Representation Neural Networks for Feature Selection,” J. Adv. Comput. Intell. Intell. Inform., Vol.3 No.4, pp. 326-331, 1999.Data files: