Turn a document term matrix into a term co-occurrence matrix

Turn a document term matrix, whose rows index documents and whose columns index terms, into a term co-occurrence matrix. A term co-occurrence matrix's rows and columns both index terms. See details, below.

Dtm2Tcm(dtm)

Arguments

dtm

A document term matrix, generally of class dgCMatrix, though other classes, such as dgTMatrix, may also work without issue.

Value

Returns a square dgCMatrix whose rows and columns both index terms. The i, j entries of this matrix represent the count of term j across documents containing term i. Note that, while square, this matrix is not symmetric.

Examples

data(nih_sample_dtm)

tcm <- Dtm2Tcm(nih_sample_dtm)