A novel kernel clustering algorithm based selective neural network ensemble model for economic forecasting

  • Authors:
  • Jian Lin;Bangzhu Zhu

  • Affiliations:
  • School of Economics and Management, Beijing University of Aeronautics and Astronautics, Beijing, China;School of Economics and Management, Beijing University of Aeronautics and Astronautics, Beijing, China and Institute of System Science and Technology, Wuyi University, Jiangmen, Guangdong, China

  • Venue:
  • ISICA'07 Proceedings of the 2nd international conference on Advances in computation and intelligence
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

In this study, a novel kernel clustering algorithm based selective neural network ensemble method, i.e. KCASNNE, is proposed. In this model, on the basis of different training subsets generated by bagging algorithm, the feature extraction technique, kernel principal component analysis (KPCA), is used to extract their data features to train individual networks. Then kernel clustering algorithm (KCA) is used to select the appropriate number of ensemble members from the available networks. Finally, the selected members are aggregated into a linear ensemble model with simple average. For illustration and testing purposes, the proposed ensemble model is applied for economic forecasting.