polmineR (version 0.6.2)

as.TermDocumentMatrix: as.TermDocumentMatrix / as.DocumentTermMatrix

Description

Method for type conversion, to generate the classes "TermDocumentMatrix" or "DocumentTermMatrix" contained in the "tm" package. The classes inherit from the "simple_triplet_matrix"-class defined in the "slam"-package. A "DocumentTermMatrix" is required as input by the "topicmodels" package, for instance.

Usage

as.TermDocumentMatrix(x, ...)
"as.TermDocumentMatrix"(x, pAttribute, sAttribute, from = NULL, to = NULL, strucs = NULL, rmBlank = TRUE, verbose = TRUE, robust = FALSE, mc = FALSE, progress = TRUE)
"as.TermDocumentMatrix"(x, col, pAttribute = NULL, verbose = TRUE)
"as.DocumentTermMatrix"(x, col)
"as.TermDocumentMatrix"(x, pAttribute = NULL, col = NULL, verbose = TRUE)
"as.DocumentTermMatrix"(x, pAttribute = NULL, col = NULL, verbose = TRUE)

Arguments

x

some object

...

to make the check happy

pAttribute

the p-attribute

sAttribute

the s-attribute

from

bla

to

bla

strucs

bla

rmBlank

bla

verbose

bla

robust

bla

mc

logical

progress

logical, whether to show progress bar

col

the column to use of assembling the matrix

Value

a TermDocumentMatrix

Details

The type conversion-method can be applied on object of the class "bundle", or classes inheriting from the "bundle" class. If counts or some other measure is present in the "stat" slots of the objects in the bundle, then the values in the column indicated by "col" will be turned into the values of the sparse matrix that is generated. A special case is the generation of the sparse matrix based on a "partitionBundle" that does not yet include counts. In this case, a "pAttribute" needs to be provided, then counting will be performed, too.

Examples

if (require(polmineR.sampleCorpus) && require(rcqp)){
   use("polmineR.sampleCorpus")
   
   # do-it-yourself 
   p <- partition("PLPRBTTXT", text_date=".*", regex=TRUE)
   pB <- partitionBundle(p, sAttribute="text_date")
   pB <- enrich(pB, pAttribute="word")
   tdm <- as.TermDocumentMatrix(pB, col="count")
   
   # leave the counting to the as.TermDocumentMatrix-method
   pB2 <- partitionBundle(p, sAttribute="text_date")
   tdm <- as.TermDocumentMatrix(pB2, pAttribute="word")
   
   # diretissima
   pB3 <- as.TermDocumentMatrix("PLPRBTTXT", pAttribute="word", sAttribute="text_date")
}