This paper pointed out the shortages of reasoning algorithms of the present belief decision trees and then proposed a new reasoning algorithm.The new algorithm makes the best of the advantage of belief function theory for handling uncertainty,revises the outputs of the relevant leaf nodes using the discounting rule,takes into account the test attribute weights,and thus improves the classification precision and reliability.Additionally,a new evidential combination rule was proposed to overcome the shortages ...
| 
推荐 
点击下列相关研究机构和相关文献作者，可以直接查到这些机构和作者被《中国知识资源总库》收录的其它文献，使您全面了解该机构和该作者的研究动态和历史。
