论文部分内容阅读
针对属性值为区间数的大量属性决策问题,提出考虑决策者风险偏好和属性约简的熵可靠性决策模型。首先引入风险偏好因子将决策者进行分类,然后提出基于熵可靠性的属性约简方法并提取有效属性,匹配相应的信息熵权重确定方法和排序方法进行方案排序和择优,通过案例分析验证该方法的科学合理性;并通过选取不同的风险偏好因子,对决策对象排序结果进行灵敏度分析。结果表明随着决策者风险偏好程度的增加,决策属性保留个数递减;最后经算法对比表明该算法考虑决策信息的可靠性,减少信息损失且计算简便。
Aiming at a large number of attribute decision making problems whose attribute value is interval number, an entropy reliability decision model considering the decision maker’s risk preference and attribute reduction is proposed. Firstly, the risk preference factor is introduced to classify the decision makers. Then, the attribute reduction method based on entropy reliability is proposed and the valid attributes are extracted. The corresponding information entropy weight determination methods and ranking methods are selected for program ranking and merit selection. The case analysis is used to verify the method The scientific rationality of the decision-making objects is also analyzed by choosing different risk preference factors. The results show that as the risk preference of decision makers increases, the number of decision attributes retained decreases. Finally, the comparison of algorithms shows that the algorithm considers the reliability of decision information, reduces the information loss and is easy to calculate.