首页> 外文会议>Youth Academic Annual Conference of Chinese Association of Automation >Learning belief regression trees from evidential data
【24h】

Learning belief regression trees from evidential data

机译:从证据数据学习信仰回归树

获取原文

摘要

Regression trees are extended to be learnt from data with epistemic uncertainty. Modelling uncertainty with belief functions, the attribute selection strategy based on error interval is discussed and a complete tree construction procedure is proposed. As a general approach, error intervals weighted by mass functions are calculated for making the best splitting choice. Including classical regression trees as a special case, belief regression trees can deal with various kinds of uncertain data such as imprecise, uncertain and noisy ones.
机译:延长回归树从认知不确定性与数据中的数据学习。讨论了基于误差间隔的属性选择策略建模不确定性,提出了一种完整的树施工过程。作为一种通用方法,计算由质量函数加权的误差间隔,用于制作最佳分割选择。包括古典回归树作为一个特例,信仰回归树可以处理各种不确定的数据,如不精确,不确定和嘈杂的数据。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号