Extract Cross-Category Similarities from Full Similarity Matrix

Given a full similarity matrix and category information, extracts pairwise similarities between a reference category and other categories into a long-format data frame suitable for visualization and analysis.

Usage

extract_cross_category_similarities(
  similarity_matrix,
  docs_data,
  reference_category,
  compare_categories = NULL,
  category_var = "category",
  id_var = "display_name",
  name_var = NULL
)

Arguments

similarity_matrix

A square similarity matrix (n x n).

docs_data

A data frame containing document metadata with at least:

category_var

Column indicating category membership

id_var

Column with unique document identifiers

reference_category

Character string specifying the reference category to compare against.

compare_categories

Character vector of categories to compare with the reference. If NULL, compares with all categories except reference.

category_var

Name of the column containing category information (default: "category").

id_var

Name of the column containing document IDs (default: "display_name").

name_var

Optional name of column with display names (default: NULL, uses id_var).

Value

A data frame with columns:

ref_id

Reference document ID

ref_name

Reference document name (if name_var provided)

other_id

Comparison document ID

other_name

Comparison document name (if name_var provided)

other_category

Category of comparison document

similarity

Cosine similarity value

Examples

# \donttest{
articles <- TextAnalysisR::SpecialEduTech[1:6, ]
articles$display_name <- paste0("d", seq_len(nrow(articles)))
term_matrix <- as.matrix(quanteda::dfm(quanteda::tokens(articles$abstract)))
normalized_matrix <- term_matrix / sqrt(rowSums(term_matrix ^ 2))
similarity_matrix <- normalized_matrix %*% t(normalized_matrix)
dimnames(similarity_matrix) <- list(articles$display_name, articles$display_name)
cross_similarities <- extract_cross_category_similarities(
  similarity_matrix  = similarity_matrix,
  docs_data          = articles,
  reference_category = "thesis",
  compare_categories = "journal_article",
  category_var       = "reference_type",
  id_var             = "display_name",
  name_var           = "title"
)
# }