Belief Network Classifier

Classification is an important research area in Artificial Intelligence, which has a broad range of applications such as pattern recognition, diagnosis, data mining, and so on. A best classifier can be built by using belief networks. This paper mainly discusses how to build the Naive Bayes classifiers, the Augmented Naive Bayes classifiers, and the General Belief Network classifiers. Their respective advantages and shortcomings also be shown by a detailed comparison.