Using Bayesian network and AIS to perform feature subset selection

  • Authors:
  • Boyun Zhang

  • Affiliations:
  • School of Information Science and Engineering, Central South University, Changsha, P.R. China and Department of computer Science, Hunan Public Security College, Changsha, P.R. China

  • Venue:
  • ICIC'09 Proceedings of the Intelligent computing 5th international conference on Emerging intelligent computing technology and applications
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

Bayesian networks have been widely applied to the feature selection problem. The proposed methods learn a Bayesian network from the available dataset and utilize the Markov Blanket of the target feature to select the relevant features. And we apply an artificial immune system as search procedure to the Bayesian network learning problem. It find suitable Bayesian networks that best fit the dataset. Due to the resulting multimodal search capability, several subsets of features are obtained. Experimental results were carried out in order to evaluate the proposed methodology in classification problems and the subsets of features were produced.