The Research and Improvement of Rough Set Based Decision Tree

摘要：

The traditional decision tree algorithm attributes selection strategy is based on information theory.The new attribute selection strategy discern value based is much more better than the traditional one,but it has to compare all objects(records),so has a higher time complexity.In this paper,we use the H-important and L-important concept to improve the new strategy,and reduce the time complexity.

关键词： Decision tree：Rough set Discern value H-important L-important

作者: Jing YANG Han WU Jianpei ZHANG

会议类型: 国际会议

会议名称: The International Conference Information Computing and Automation(2007国际信息计算与自动化会议)

会议地点: 成都

会议语种:英文

页码: 1081-1084

在线出版日期: 2007-12-19（万方平台首次上网日期，不代表论文的发表时间）