smith_waterman_pairwise {text.alignment}

Perform multiple alignments using Smith-Waterman

Description

Utility function to perform all pairwise combinations of alignments between text.

Usage

smith_waterman_pairwise(a, b, FUN = identity, ...)

Arguments

a

a data.frame with columns doc_id and text. Or a character vector where the names of the character vector respresent a doc_id and the character vector corresponds to the text.

b

a data.frame with columns doc_id and text. Or a character vector where the names of the character vector respresent a doc_id and the character vector corresponds to the text.

FUN

a function to apply on an object of class smith_waterman which has done the pairwise alignment. Defaults to identity. Other options are as.data.frame or your own function. See the examples.

...

other arguments passed on to smith_waterman

Value

a list of pairwise Smith-Waterman comparisons after which the FUN argument is applied on all of these pairwise alignments. The output of the result of FUN is enriched by adding a list element a_doc_id and b_doc_id which correspond to the doc_id's provided in a and b and which can be used in order to identify the match.

See Also

smith_waterman

Examples

x <- data.frame(doc_id = c(1, 2),
                text = c("This is some text", "Another set of texts."),
                stringsAsFactors = FALSE)
y <- data.frame(doc_id = c(1, 2, 3),
                text = c("were as some thing", "else, another set", NA_character_),
                stringsAsFactors = FALSE)
alignments <- smith_waterman_pairwise(x, y)
alignments
alignments <- smith_waterman_pairwise(x, y, FUN = as.data.frame)
do.call(rbind, alignments)
alignments <- smith_waterman_pairwise(x, y, 
                                      FUN = function(x) list(sim = x$similarity))
do.call(rbind, alignments)

x <- c("1" = "This is some text", "2" = "Another set of texts.")
y <- c("1" = "were as some thing", "2" = "else, another set", "3" = NA_character_)
alignments <- smith_waterman_pairwise(x, y)