基于奇异值分解的多评价结论集结方法-JournalofNortheastern.PDF

第28 卷第2期 东 北 大 学 学 报 ( 自 然 科 学 版 ) Vol128, No. 2 2 00 7 年 2 月 Journal of Northeastern U niversit ( Natural Science) Feb. 2 0 0 7 郭亚军, 易平涛 ( , 110004) : / 0 , , # , , # , # # : ; ; ; ; : C 931 : A : 1005-3026( 2007) 02-0278-04 A Mult-i evaluation Conclusion Aggregation Method Based on Singular Value Decomposition G UO Ya-j un, YI Ping-tao ( School of Business Administration, Northeastern Universit , Shen ang 110004, China. Correspondent : GUO Y a- jun, professor, E-mail: jguo200 1 @ 163. com) Abstract: A mult-i evaluation conclusion aggregation method based on SVD ( singular value decomposition) is put forward aiming at the inconsistenc betw een the evaluation conclusions draw n b different methods. In this w a , the SVD is done for the numerical matrix composed of multiple inconsistenc conclusions. Then, the number of singular values to be kept are determined b equilibrating the tw o defined performance indices, i. e. , consistenc and reliabilit . Based on the principle of singular value decomposition, an approximate conclusion matrix is given as the optimization of original conclusions. The method has advantages of picking up common information from mult-i evaluation, w eakening the impact due to extreme evaluation conclusions and controlling flex ibl the aggregation process. A numerical ex ample is given to verif the validit of the method. Key words: comprehensive evaluation; singular value decomposition ( SVD ) ; combined evaluation; consistenc ; reliabilit ( )