The method returns a data.table with the number of matches of a query
or multiple queries (optionally frequencies) in a corpus or subcorpus as
partitioned by one or two s-attributes.
dispersion(.Object, ...)
# S4 method for class 'slice'
dispersion(
.Object,
query,
s_attribute,
cqp = FALSE,
p_attribute = getOption("polmineR.p_attribute"),
freq = FALSE,
fill = TRUE,
mc = FALSE,
progress = FALSE,
verbose = FALSE,
...
)
# S4 method for class 'partition'
dispersion(
.Object,
query,
s_attribute,
cqp = FALSE,
p_attribute = getOption("polmineR.p_attribute"),
freq = FALSE,
fill = TRUE,
mc = FALSE,
progress = TRUE,
verbose = FALSE,
...
)
# S4 method for class 'subcorpus'
dispersion(
.Object,
query,
s_attribute,
cqp = FALSE,
p_attribute = getOption("polmineR.p_attribute"),
freq = FALSE,
fill = FALSE,
mc = FALSE,
progress = TRUE,
verbose = FALSE,
...
)
# S4 method for class 'corpus'
dispersion(
.Object,
query,
s_attribute,
cqp = is.cqp,
p_attribute = getOption("polmineR.p_attribute"),
freq = FALSE,
fill = TRUE,
mc = FALSE,
progress = FALSE,
verbose = FALSE,
...
)
# S4 method for class 'character'
dispersion(
.Object,
query,
s_attribute,
cqp = is.cqp,
p_attribute = getOption("polmineR.p_attribute"),
freq = FALSE,
fill = TRUE,
mc = FALSE,
progress = TRUE,
verbose = TRUE,
...
)
# S4 method for class 'hits'
dispersion(
.Object,
source,
s_attribute,
freq = FALSE,
fill = TRUE,
verbose = TRUE,
...
)
# S4 method for class 'remote_corpus'
dispersion(.Object, ...)
# S4 method for class 'remote_subcorpus'
dispersion(.Object, ...)A corpus, subcorpus or partition object
or a corpus provided by a character string.
Further parameters.
A character vector stating one or multiple queries.
A character vector (length 1 or 2) providing s-attributes.
If logical, whether the query is a CQP query, if it is a
function that is passed in, the function will be applied to the query to
guess whether query is a CQP query
Length one character vector, the p-attribute that
will be looked up (typically 'word' or 'lemma').
A logical value, whether to calculate normalized frequencies.
A logical value, whether to report zero matches. Defaults
to TRUE. But note that if there are few matches and many values of
the s-attribute(s), the resulting data structure is sparse and potentially
bloated.
A logical value, whether to use multicore.
A logical value, whether to show progress.
A logical value, whether to be verbose.
The source of the evaluation the hits reported in
.Object are based on, a corpus, subcorpus or
partition object.
A data.table.
A data.table.
Augmenting the data.table with zeros for subcorpora that do
not yield query matches (argument fill = TRUE) may require
adding many new columns. A respective warning issued by the
data.table package is supplemented an additional explanatory note
of the polmineR package.
The worker behind the dispersion-method is the hits-method.
count
use("polmineR")
#> ✔ corpus loaded: GERMAPARLMINI (version: 0.1.0 | build date: 2023-04-16)
dispersion("GERMAPARLMINI", query = "Integration", s_attribute = "date")
#> query date count
#> <char> <char> <int>
#> 1: Integration 2009-10-27 1
#> 2: Integration 2009-10-28 0
#> 3: Integration 2009-11-10 7
#> 4: Integration 2009-11-11 15
#> 5: Integration 2009-11-12 0
test <- partition("GERMAPARLMINI", date = ".*", p_attribute = NULL, regex = TRUE)
#> ... get cpos and strucs
integration <- dispersion(
test, query = "Integration",
p_attribute = "word", s_attribute = "date"
)
integration <- dispersion(test, "Integration", s_attribute = c("date", "party"))
integration <- dispersion(test, '"Integration.*"', s_attribute = "date", cqp = TRUE)