Bundle Class

A bundle is used to combine several objects (partition, context, features, cooccurrences objects) into one S4 class object. Typically, a class inheriting from the bundle superclass will be used. When working with a context_bundle, a features_bundle, a cooccurrences_bundle, or a context_bundle, a similar set of standard methods is available to perform transformations.

# S4 method for bundle
name(x) <- value

# S4 method for bundle
length(x)

# S4 method for bundle
names(x)

# S4 method for bundle,vector
names(x) <- value

# S4 method for bundle
unique(x)

# S4 method for bundle,bundle
+(e1, e2)

# S4 method for bundle,textstat
+(e1, e2)

# S4 method for bundle
[[(x, i)

# S4 method for bundle
[[(x, i) <- value

# S4 method for bundle
$(x, name)

# S4 method for bundle
$(x, name) <- value

# S4 method for bundle
sample(x, size)

# S4 method for list
as.bundle(object, ...)

# S4 method for textstat
as.bundle(object)

# S3 method for bundle
as.data.table(x, keep.rownames, col, ...)

# S4 method for bundle
as.matrix(x, col)

# S4 method for bundle
subset(x, ...)

# S4 method for bundle
as.list(x)

# S3 method for bundle
as.list(x, ...)

# S4 method for bundle
get_corpus(x)

Arguments

x	a bundle object
value	character string with a name to be assigned
e1	object 1
e2	object 2
i	An integer value to index a bundle object.
name	The name of an object in the bundle object.
size	number of items to choose to generate a sample
object	A bundle object.
...	Further parameters
keep.rownames	Required argument to safeguard consistency with S3 method definition in the data.table package. Unused in this context.
col	columns of the data.table to use to generate an object.

Slots

corpus

The CWB corpus the objects in the bundle are based on, a length 1 character vector.

objects

An object of class list.

p_attribute

Object of class character.

encoding

The encoding of the corpus.

Author

Andreas Blaette

Examples

parties <- s_attributes("GERMAPARLMINI", "party")
parties <- parties[-which(parties == "NA")]
party_bundle <- partition_bundle("GERMAPARLMINI", s_attribute = "party")
length(party_bundle)
#> [1] 6
names(party_bundle)
#> [1] "B90_DIE_GRUENEN" "CDU_CSU"         "DIE_LINKE"       "FDP"            
#> [5] "NA"              "SPD"            
get_corpus(party_bundle)
#> [1] "GERMAPARLMINI"
party_bundle <- enrich(party_bundle, p_attribute = "word")
summary(party_bundle)
#>              name  size p_attribute unique
#> 1 B90_DIE_GRUENEN 23052        word   4186
#> 2         CDU_CSU 66490        word   8486
#> 3       DIE_LINKE 22542        word   4531
#> 4             FDP 31505        word   4930
#> 5              NA 31310        word   5032
#> 6             SPD 47302        word   6692
parties_big <- party_bundle[[c("CDU_CSU", "SPD")]]
summary(parties_big)
#>      name  size p_attribute unique
#> 1 CDU_CSU 66490        word   8486
#> 2     SPD 47302        word   6692
p <- partition("GERMAPARLMINI", date = "2009-11-11")
#> ... get encoding: latin1
#> ... get cpos and strucs
pb <- partition_bundle(p, s_attribute = "party")
names(pb)
#> [1] "B90_DIE_GRUENEN" "CDU_CSU"         "DIE_LINKE"       "FDP"            
#> [5] "NA"              "SPD"            
pb[["NA"]] <- NULL
names(pb)
#> [1] "B90_DIE_GRUENEN" "CDU_CSU"         "DIE_LINKE"       "FDP"            
#> [5] "SPD"            
pb <- partition_bundle("GERMAPARLMINI", s_attribute = "party")
pb$SPD # access partition names "SPD" in partition_bundle pb
#> ** partition object **
#> corpus:             GERMAPARLMINI
#> name:               SPD
#> s-attributes:       party = SPD
#> cpos:                
#> 908 pairs of corpus positions
#> size:                
#> 47302 tokens
#> count:               
#> not available
pb <- partition_bundle("GERMAPARLMINI", s_attribute = "party")
pb$"NA" <- NULL # quotation needed if name is "NA"
use("polmineR")
#> ... activating corpus: GERMAPARLMINI (version: 0.0.1 | build date: 2019-02-23)
#> ... activating corpus: REUTERS
pb <- partition_bundle("REUTERS", s_attribute = "id")
coocs <- cooccurrences(pb, query = "oil", cqp = FALSE)
#> Warning: No hits for query oil (returning NULL)
dt <- as.data.table(coocs, col = "ll")
m <- as.matrix(dt[, 2:ncol(dt)], rownames = dt[["token"]])