本文已被:浏览 1276次 下载 681次
投稿时间:2017-03-09
投稿时间:2017-03-09
中文摘要: 针对目前海量数据分析较多情况下从传统的单条记录转变为一个区间对象,将决策粗糙集中风险的概念引入至区间值决策表中,给出了区间值决策表决策风险的定义,并提出了决策风险最小化的属性约简方法.该方法可以保证所得到的约简集合相对于决策属性具有较强的分类能力,同时保证约简集合的决策风险最小.区间值决策表的决策风险最小化约简使得定义的约简具有更强的理论性和可解释性.
Abstract:It is found that it is more reasonable when the study objects are treated as intervals than as single records in many circumstances.An attribute reduction method in interval-valued decision tables is proposed.The concept of risk in decision-theoretic rough sets is introduced into interval-valued decision tables,and the definition of decision risk is given.The corresponding reduction method is also proposed,which can not only guarantee that the obtained reduction set has stronger classification ability relative to the decision attributes,but also make sure the minimum decision risk.The defined reduction of decision risk in interval-valued decision tables is more theore-tical and interpretable.
keywords: interval-valued decision tables decision-theoretic rough sets minimum risk attribute reduction
文章编号:20175012 中图分类号: 文献标志码:
基金项目:国家自然科学基金(61272437,61305094);上海市教育发展基金会和上海市教育委员会"晨光计划"(13CG58);上海市自然科学基金(13ZR1417500).
作者 | 单位 | |
徐菲菲 | 上海电力学院 计算机科学与技术学院 | xufeifei1983@hotmail.com |
Author Name | Affiliation | |
XU Feifei | School of Computer Science and Technology, Shanghai University of Electric Power, Shanghai 200090, China | xufeifei1983@hotmail.com |
引用文本:
徐菲菲.区间值决策表的决策风险最小化属性约简[J].上海电力大学学报,2017,33(5):471-476.
XU Feifei.Attribute Reduction Based on Minimum Decision Riskin Interval-Valued Decision Tables[J].Journal of Shanghai University of Electric Power,2017,33(5):471-476.
徐菲菲.区间值决策表的决策风险最小化属性约简[J].上海电力大学学报,2017,33(5):471-476.
XU Feifei.Attribute Reduction Based on Minimum Decision Riskin Interval-Valued Decision Tables[J].Journal of Shanghai University of Electric Power,2017,33(5):471-476.