total_correlation: Estimation of total correlation for a set of categorical...

In msu: Multivariate Symmetric Uncertainty and Other Measurements

Description

Total Correlation is a generalization of information gain (IG) to measure the dependency of a set of categorical random variables (see https://en.wikipedia.org/wiki/Total_correlation).

Usage

total_correlation(table_variables, table_class)

C(table_variables, table_class)

Arguments

table_variables	A list of factors as categorical variables.
table_class	A factor representing the class of the case.

Value

Total correlation estimation for the variable set table.variables, table.class.

Examples

total_correlation(list(factor(c(0,1)), factor(c(1,0))), factor(c(0,0)))
total_correlation(list(factor(c('a','b')), factor(c('a','b'))),
    factor(c('a','b')))
## Not run: 
total_correlation(list(factor(c(0,1)), factor(c(1,0))), c(0,0))
total_correlation(c(factor(c(0,1)), factor(c(1,0))), c(0,0))

## End(Not run)

msu documentation built on May 2, 2019, 6:43 a.m.