摘要
稳定性是评估分类算法的一个重要方面.本文介绍了贝叶斯分类方法以及两种度量算法稳定性的方法,实验研究了几种流行的分类算法的稳定性.研究的目的是确定树增强的朴素贝叶斯网络分类方法的稳定性.实验结果表明,树增强的朴素贝叶斯网络分类方法是稳定的.
The stability is an important criterion of evaluating classification algorithms. Bayesian network classification method and TAN model arc firstly introduced in this paper. An empirical investigation, which compares the stability of several typical classification approaches(decision tree, Naive Bayes)with TAN by utilizing two measure methods, is detailedly described. The purpose of the study is to determine the stability of the TAN classifier. Experimental results show that Tree Augmented Naive Bayes network classifier is stable.
出处
《模式识别与人工智能》
EI
CSCD
北大核心
2004年第3期275-280,共6页
Pattern Recognition and Artificial Intelligence
关键词
贝叶斯网络
稳定性
差异
Bayesian Network
Stability
Variance
