An Improved Weighted Mixed Bayes Classification Model
Wei Xiang Xu, Ling Sun, Xu Min Liu · Advanced materials research · 2011
The paper investigates kinds of classical Bayes classification arithmetic and summarizes their advantage and disadvantage, then introduces an improved weighted mixed Bayes classification model. It divides attribute sets into several subsets theorem. The subsets are trained by TAN (Tree Augmented Naive Bayes) and the results are integrated by weighted formula. At the same, the paper introduced a new method to compute the weights of attribute subsets. Finally, Experimental results show that this model has higher classification accuracy and practicability．