Article
Merging subsets of attributes to improve a hybrid consistency-based filter: a case of study in product unit neural networks
Author/s | Tallón Ballesteros, Antonio Javier
Riquelme Santos, José Cristóbal Ruiz Sánchez, Roberto |
Department | Universidad de Sevilla. Departamento de Lenguajes y Sistemas Informáticos |
Publication Date | 2016 |
Deposit Date | 2023-05-08 |
Published in |
|
Abstract | This paper presents a quality enhancement of the selected features by a hybrid filter-based jointly on feature ranking and feature subset selection (FR-FSS) using a consistency-based measure via merging new features which ... This paper presents a quality enhancement of the selected features by a hybrid filter-based jointly on feature ranking and feature subset selection (FR-FSS) using a consistency-based measure via merging new features which are obtained applying other FR-FSS evaluated with a correlation metric. The goal is to overcome the accuracy of a neural network classifier containing product units as hidden nodes combined with a feature selection pre-processing step by means of a single consistency-based FR-FSS filter. Neural models are trained with a refined evolutionary programming approach called two-stage evolutionary algorithm. The experimentation has been carried out in eight complex classification problems, seven out of them from UCI (University of California at Irvine) repository and one real-world prob lem, with high test error rates (around 20%) with powerful classifiers such as 1-nearest neighbour or C4.5. Non-parametric statistical tests revealed that the new proposal significantly improves the accuracy of the neural models. |
Funding agencies | Comisión Interministerial de Ciencia y Tecnología (CICYT). España Junta de Andalucía |
Project ID. | IN2011-28956-C02-02
TIN2014-55894-C2-R P11-TIC-7528 |
Citation | Tallón Ballesteros, A.J., Riquelme Santos, J.C. y Ruiz Sánchez, R. (2016). Merging subsets of attributes to improve a hybrid consistency-based filter: a case of study in product unit neural networks. Connection Science, 28 (3), 242-257. https://doi.org/10.1080/09540091.2016.1149146. |
Files | Size | Format | View | Description |
---|---|---|---|---|
Merging subsets of attributes ... | 1.255Mb | [PDF] | View/ | |