Small sample efficiency performance-based clustering method

Abstract

The invention discloses a small sample efficiency performance-based clustering method. The method includes the following steps that: an optimal leading surface-based super efficiency model is established; a worst leading surface-based super efficiency model is established; and decision-making units are classified through adopting Ward's method. With the small sample efficiency performance-based clustering method of the invention, the clustering problem of multiple-input-multiple-output small samples is better solved, and a defect that a traditional DEA method generally only classifies decision-making units as effective units and inactive units can be eliminated. According to the data envelopment analysis and data mining technology decision-decision-making unit discriminant analysis method,the efficiency values of the super-efficiency DEA are adopted to perform clustering analysis on the decision-making units, and therefore, the advantage of no need for dimension processing, of data envelopment analysis, is reserved, and reasonable clustering results can be obtained.

Description

technical field [0001] The invention relates to a clustering method based on small sample efficiency performance. Background technique [0002] Cluster analysis is a process of classifying data into different classes or clusters, so objects in the same cluster have great similarity, while objects in different clusters have great dissimilarity. Cluster analysis is a method of simplifying data through data modeling. Traditional statistical clustering analysis methods include systematic clustering, decomposition, joining, dynamic clustering, ordered sample clustering, overlapping clustering and fuzzy clustering, etc. [0003] The above traditional clustering analysis methods have the following two problems: 1. The above method mainly clusters samples by constructing functions, but it is obviously not feasible to cluster samples with multiple output types; 2. Using these methods Statistical analysis needs to be carried out on a large number of samples, and when the samples are...

AI Technical Summary

Problems solved by technology

[0003] The above traditional clustering analysis methods have the following two problems: 1. The above method mainly clusters samples by constructing functions, but it is obviously not feasible to cluster samples with multiple output types; 2. Using these methods Statistical analysis needs to be performed on a large number of samples. When the sample is small, the above method will have a large deviation

However, when the output of a decision-making unit is greater than the output of all other decision-making units (input-type super-efficiency model) or the input is smaller than the input of all other decision-making units (output-type super-efficiency model), the corresponding super-efficiency model will be There is no solution

Images