Post-processing a classifier's predictions: Strategies and empirical evaluation
暂无分享,去创建一个
In this paper, we propose an approach allowing to revise the outputs of a classifier in order to take into account the available domain knowledge. This approach can be applied for any classifier be it probabilistic or not. We propose post-processing criteria and methods to encode and exploit different kinds of domain knowledge. Finally, we provide experimental studies on a set of benchmarks.
[1] Nir Friedman,et al. Bayesian Network Classifiers , 1997, Machine Learning.
[2] Habiba Drias,et al. An intrusion detection and alert correlation approach based on revising probabilistic classifiers using expert knowledge , 2012, Applied Intelligence.
[3] J. Ross Quinlan,et al. Induction of Decision Trees , 1986, Machine Learning.