26 search results - page 3 / 6 » A Bayesian network approach to feature selection in mass spe... |
KDD
1998
ACM
13 years 9 months ago
1998
ACM
BAYDA is a software package for flexible data analysis in predictive data mining tasks. The mathematical model underlying the program is based on a simple Bayesian network, the Na...
ESANN
2006
13 years 6 months ago
2006
In this article, we extend a local prototype-based learning model by active learning, which gives the learner the capability to select training samples during the model adaptation...
AAAI
2008
13 years 7 months ago
2008
Based on Information Theory, optimal feature selection should be carried out by searching Markov blankets. In this paper, we formally analyze the current Markov blanket discovery ...
PRL
2000
13 years 5 months ago
2000
We demonstrate the advantages of using Bayesian multi layer perceptron (MLP) neural networks for image analysis. The Bayesian approach provides consistent way to do inference by c...
SSPR
2004
Springer
13 years 10 months ago
2004
Springer
We focus on characterizing spatial region data when distinct classes of structural patterns are present. We propose a novel statistical approach based on a supervised framework for...