区间值决策表中基于证据理论的测试代价敏感属性约简
Test Cost-Sensitive Attribute Reduction in Interval-Valued Decision Tables Based on Evidence Theory
范译文 1廖淑娇 1吴迪1
作者信息
- 1. 闽南师范大学数学与统计学院,福建 漳州 363000;闽南师范大学福建省粒计算及其应用重点实验室,福建 漳州 363000
- 折叠
摘要
属性约简一直都是区间值信息系统的主要研究内容,Dempster-Shafer证据理论应用信任函数和似然函数解决不确定性问题.根据证据理论相关知识研讨区间值决策表中测试代价敏感的属性约简方法,建立理论模型,并分别基于信任函数和似然函数提出启发式的属性约简算法.经实例检验,所提出算法的效果良好.
Abstract
Attribute reduction has always been the main research content of interval-valued information systems.Dempster-Shafer evidence theory applies belief function and plausibility function to solve uncertainty problems.In this paper,we study the test cost-sensitive attribute reduction method in interval-valued decision tables based on the relevant knowledge of evidence theory.The theoretical model is built,and heuristic attribute reduction algorithms are proposed based on belief function and plausibility function,respectively.The effectiveness of the proposed algorithms is verified through examples.
关键词
区间值/决策表/属性约简/测试代价/证据理论Key words
interval value/decision table/attribute reduction/test cost/evidence theory引用本文复制引用
出版年
2024
NETL
NSTL
万方数据