Mixed weight kernel entropy component analysis-based enterprise supplier evaluation method
The invention discloses a mixed weight kernel entropy component analysis-based enterprise supplier evaluation method. The method is characterized by comprising the following steps of (1) determining supplier evaluation criteria, and performing output and input index classification; (2) taking a reci...
Saved in:
Main Authors | , , , , , , |
---|---|
Format | Patent |
Language | Chinese English |
Published |
15.09.2017
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | The invention discloses a mixed weight kernel entropy component analysis-based enterprise supplier evaluation method. The method is characterized by comprising the following steps of (1) determining supplier evaluation criteria, and performing output and input index classification; (2) taking a reciprocal of an input index obtained in the step (1) and performing normalization processing on all indexes by utilizing a maximum and minimum method to enable a range to be [0,1]; (3) performing kernel entropy component analysis on data obtained in the step (2) to calculate out a kernel matrix K corresponding to a sample matrix X; (4) decomposing the kernel matrix obtained in the step (3), working out eigenvalues and eigenvectors, and selecting out maximal first d eigenvalues corresponding to values defined in the specification and corresponding eigenvectors to perform projection; and (5) calculating weight vectors of corresponding new base coordinates, and performing weighted summation with projection values obtaine |
---|---|
Bibliography: | Application Number: CN20171571455 |