基于保护少数人意见的群体一致性分析  被引量:8

Group consistency analysis for protecting the minority views

在线阅读下载全文

作  者:熊才权[1] 李德华[1] 金良海[1] 

机构地区:[1]华中科技大学图像识别与人工智能研究所

出  处:《系统工程理论与实践》2008年第10期102-107,共6页Systems Engineering-Theory & Practice

基  金:国家自然科学基金(69775022);湖北省自然科学基金(2007ABA025)

摘  要:在群体研讨过程中,少数人意见是十分重要的.提出了一种可以保护少数意见的决策共识达成模型及研讨反馈机制.该方法先根据专家意见对专家群体进行聚类分析,得到专家子群体簇,如果某一子群体所含专家人数较少,则该子群体的意见称为少数人意见,特别地,如果在聚类中出现奇异点,则该奇异点的专家意见称为个别人意见,然后主持人根据少数人成员的特性将少数人意见单独提出,使专家群体参考少数人意见重新研讨并给出新的偏好矢量,重复这一过程直到群体接受或摈弃少数人意见,使专家意见得到统一.最后用一个实例说明了该方法的可行性.In the process of consensus building, the minority views are very important. This paper proposes a consensus model and a discussion mechanism that can protect the minority views automatically. The proposed method first clusters the expert group into a set of subgroups by a threshold value of similarity of pairs of expert preference vectors. If a subgroup includes a small number of experts, the opinion of the subgroup is called minority view. Then the mediator will put forward the minority views to the group according to the characteristics of the minority, and urge the group to discuss again referencing the minority views and submit some new preferences. The process is repeated until the group reaches a consensus after considering or eliminating the minority views. Finally, the paper gives an example to illustrate the feasibility of the method.

关 键 词:少数人意见 专家聚类 群体一致性 

分 类 号:TP391[自动化与计算机技术—计算机应用技术]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象