利用类概率估计的平均树增强朴素贝叶斯网络结构

树增强朴素贝叶斯模型通过放松条件属性独立来改进贝叶斯模型，结构学习效率较高且简单。然而在一些实际试验测试中，树增强朴素贝叶斯分类模型的分类精确性和失误率的效果却不好。因此在本文中，设计了平均的树增强朴素贝叶斯分类算法来改进分类的效果，并且利用条件对数似然来测试分类估计的效果，最后利用Weka平台公布的大量的UCI数据集进行试验，结果表明平均树增强朴素贝叶斯分类模型明显优于树增强的朴素贝叶斯分类模型。

Improving Tree augmented Naive Bayes based on class probability estimation

Tree Augmented Naive Bayes (TAN) improves Naive Bayes (NB) by weakening its conditional attribute independence assumption, while maintaining efficiency and simplicity. In many real-world applications, however, classification accuracy or error rate is not enough. Thus, in this paper, we investigate Averaged Tree Augmented Naive Bayes (ATAN) algorithm to improve its class probability estimation performance .And we estimate performance of ATAN in terms of conditional log likelihood (CLL) .The experimental results on a large number of UCI datasets published on the main web site of Weka platform show that ATAN significantly outperforms TAN and all the other algorithms used to compare in terms of CLL.