Low dimensional successive projection pursuit cluster (LDSPPC) model comprehensive evaluation method, device and application

Abstract

The invention discloses a low dimensional successive projection pursuit cluster (LDSPPC) model comprehensive evaluation method, device and application. The method comprises a step of carrying out normalized preprocessing on sample data of multiple candidate objects, and constructing a one-dimensional projection pursuit cluster model with 2 to 4 mutually orthogonal projection vectors for the candidate objects, and a step of carrying out projection pursuit cluster model vector synthesis of all dimensions of the multiple candidate objects to form a comprehensive projection pursuit cluster model, and obtaining an evaluation index importance ranking list and a candidate object quality ranking list. The swarm search intelligent algorithm of the invention has the characteristics of a fast convergence speed, convergence to a globally optimal solution and high reliability, the vector synthesis of multiple successive projection pursuit vectors is carried out, the quality of the candidate objects can be quickly evaluated, and the accuracy of candidate object quality evaluation is improved.

Description

technical field [0001] The invention relates to the technical field of computer applications, in particular to a comprehensive evaluation method, device and application based on a low-dimensional successive projection-pursuit clustering model (LDSPPC). Background technique [0002] Supplier selection and evaluation involves data processing with nonlinear and non-normal distribution of multi-indicator (high-dimensional) attributes, and the effect of conventional modeling methods is poor. The one-dimensional Projection Pursuit Clustering (PPC) model proposed by Friedman et al. in 1974 has been widely used in many fields and achieved certain results. However, for supplier selection and evaluation problems with multiple attributes and few samples, it is often difficult to choose a suitable supplier due to the insufficient information of the mined sample data and the result that multiple suppliers have the same score. [0003] In the LDSPPC modeling process, it is first necessar...

AI Technical Summary

Problems solved by technology

Gong Yan et al. (2007) used the maximum relative information entropy as the objective function to establish a 9-dimensional projection pursuit clustering model, but there is no actual case data to verify the reliability of the results

None of these papers discussed how to realize the synthesis of multiple projection pursuit vectors, which is not conducive to fully mining the sample data information for classification and sorting research

[0006] There is no software found in China that can provide LDSPPC modeling, and the commercialized DPS software developed by Tang Qiyi (2013) cannot obtain reliable results for the PPC modeling program

Since the LDSPPC model is a high-dimensional nonlinear optimization problem containing both equality and inequality constraints, it is very difficult to solve

Images