hierarchical_coverage_regex: Hierarchical Coverage of Regexes

In trinker/termco: Counts of Terms and Substrings

Description

The unique coverage of a text vector by a regex after partitioning out the elements matched by previous regexes.

Usage

hierarchical_coverage_regex(
  text.var,
  term.list,
  ignore.case = TRUE,
  sort = FALSE,
  verbose = TRUE,
  ...
)

Arguments

text.var	A text vector (vector of strings).
term.list	A list of named character vectors to match against x.
ignore.case	logical. Should case be ignored in matching the terms against x?
sort	logical. If TRUE the output is sorted by highest unique gain. If FALSE order of term input is retained.
verbose	If TRUE each iteration of the for loop prints i of n.
...	ignored.

Value

Returns a data.frame with 7 columns:

step

the order in which the regex was searched for

name

the human readable name of the bound regex group

unique_prop

the unique prop coverage of the regex

unique_n

the unique n coverage of the regex

cum_prop

the cumulative prop coverage of the regex

cum_n

the cumulative n coverage of the regex

regex

the bound (|) regex that corresponds to name

See Also

Other hierarchical_coverage functions: hierarchical_coverage_term()

Examples

regs <- setNames(
    list(c('(?i)sam', "(?i)\\bam"), '^I', '(?i)(do|will) not', '(?i)(do|will)'),
    c('am', 'I', "won't")
)
(out <- hierarchical_coverage_regex(sam_i_am, regs, ignore.case=FALSE))
summary(out)
plot(out)
plot(out, mark.one = TRUE)

# Use unnamed vectors for `term.list` too
hierarchical_coverage_regex(sam_i_am, unlist(regs, use.names = FALSE), ignore.case=FALSE)

trinker/termco documentation built on Jan. 7, 2022, 3:32 a.m.