多目标决策下的数据挖掘技术分析  

Analysis of Data Mining Technology under Multi—objective Decision Making

在线阅读下载全文

作  者:张晓川[1] Zhang Xiaochuan(China mobile Communications Group Guangdong Co.LTD.Guangzhou guangdong,510000 China)

机构地区:[1]中国移动通信集团广东有限公司,广东广州510000

出  处:《现代科学仪器》2020年第2期149-152,156,共5页Modern Scientific Instruments

摘  要:基于传统AHP多目标决策算法中,采用专家决策易受专家经验,研究领域的影响,导致预测结果偏离实际结果.在传统算法的基础上,将专家经验与多目标决策相结合,建立基于多目标决策的数据挖掘评估方法.根据目标问题建立层级结构的初始判断矩阵,结合专家打分确立指标间重要性评分,并构建群决策判断矩阵对专家评分指标进行专家权重计算,降低算法的人为影响度,通过引入最小二乘法进行判断矩阵的正互反性修正,避免矩阵的正互反性限制,最后对求解权重向量进行一致性检验,进一步来优化判断决策矩阵.实例验证表明:改进后的群决策多目标算法具备更好的方案区分度,能有效增加最终决策的整体满意度.Based on AHP traditional multi objective decision-making algorithm,the use of expert decision making is vulnerable to expert experience and the influence of research field,which leads to the deviation of prediction results from actual results.Based on the traditional algorithm,the expert experience and multi objective decision are combined to establish a data mining evaluation method based on multi-objective decision.The initial judgment matrix of hierarchical structure is established according to the target problem,and the importance score between indexes is established by combining with the expert score,and the expert weight is calculated by constructing the group decision judgment matrix to reduce the artificial influence degree of the algorithm.In the end,the consistency test is carried out to optimize the decision matrix.The example shows that the improved multi-obiective algorithm of group decision has better scheme differentiation and can effectively increase the overall satisfaction of final decision.

关 键 词:多目标决策 数据挖掘 判断矩阵 最小二乘法 

分 类 号:TP3[自动化与计算机技术—计算机科学与技术]

 

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