Attribute Suggestions

Usage

attribute_suggestion(values, type = class(values), n = 3, floor = 10)

Arguments

values: a vector of values, most probably a vector corresponding to one of the item data frame columns
type: an optional character vector, the type (class) of values. If not provided, class(values) will be used
n: the desired number of suggestions. This applies only on character, factor, numeric & integer types
floor: a numeric value. This applies only for numeric & integer types. If none of the value frequency (in %) reaches floor, then basic suggestions are returned (see details)

Value

a list of suggestions This list depends on the type of values

character, factor: the most frequent values ; list(value1 = frequency, value_2, = frequency)
numeric, integer: the most frequent values ; list(value1 = frequency, value_2, = frequency), or a standard list ; list(min, max, mean, median)
logical: the frequency for both values ; list(true = frequency, false = frequency)
Date, POSIXct: a standard list ; list(min, max)

Details

For numeric & integer values, the default strategy is to compute the n most frequent occurrences. If no occurrence reaches the floor level (% of all items), then the standard list is returned.

Examples

foo <- c(rep("banana", 5), rep("mango", 3), rep("orange", 2))
attribute_suggestion(values = foo, type = "character")
#> $banana
#> [1] 50
#> 
#> $mango
#> [1] 30
#> 
#> $orange
#> [1] 20
#>