缺损评分矩阵的论文排名

传统的竞赛评判由于受各种已知或者未知因素的影响,经常会导致数据缺损的情况出现。本文从分析评委的判分与选手的真实排名存在一个高度的线性相关入手,采用了线性回归的方法填补了缺损的数据,使之成为一个完整的数据集,再采用群组特征根法对数学建模竞赛的论文进行排序,最后用计算机模拟验证这种算法较缺损数据简单求和的算法优秀。

Rank of an imperfect data matrix of the papers

Through hypothetical test,it could be found that there was a highly linear correlation between the marks the judges given and the order of the papers.In order to make a complete data set,the method of linear regression was adopted to fill the imperfect data set,and the method of eigenvalue of group decision was taken to put the papers in order.The computer simulation results show that the method is effective.