Skip to main content

Table 3 Quantification of the characteristics of ‘Wine quality_red’ dataset

From: Empirical study of seven data mining algorithms on different characteristics of datasets for biomedical classification applications

Quantification index

Values

Sample size

1599

Number of attributes

11

Number of missing values

0

Number of classes

6

Sample size of the largest class

681

Sample size of the least class

10

Correlation coefficients1a

0.4762

Correlation coefficients2a

− 0.6830

Class entropy of task variable

0.5145

Ratio of sample size of the largest class to the least class

68.10

  1. aCorrelation coefficients1 represents the maximum of correlation coefficient between task variable and other non-task attribute variables; correlation coefficients2 represents the maximum of correlation coefficient between each pair of non-task attribute variables