tab_similarity: Tabulate cosine similarity/distance of word pairs.

In PsychWordVec: Word Embedding Research Framework for Psychological Science

Tabulate cosine similarity/distance of word pairs.

Description

Tabulate cosine similarity/distance of word pairs.

Usage

tab_similarity(
  data,
  words = NULL,
  pattern = NULL,
  words1 = NULL,
  words2 = NULL,
  unique = FALSE,
  distance = FALSE
)

Arguments

data	A wordvec (data.table) or embed (matrix), see data_wordvec_load().
words	[Option 1] Character string(s).
pattern	[Option 2] Regular expression. If words and pattern are not specified, all words in the data will be extracted.
words1, words2	[Option 3] Two sets of words for only n1 * n2 word pairs. See examples.
unique	Return unique word pairs (TRUE) or all pairs with duplicates (FALSE; default).
distance	Compute cosine distance instead? Defaults to FALSE (cosine similarity).

Value

A data.table of words, word pairs, and their cosine similarity (cos_sim) or cosine distance (cos_dist).

Download

Download pre-trained word vectors data (.RData): https://psychbruce.github.io/WordVector_RData.pdf

See Also

cosine_similarity()

pair_similarity()

plot_similarity()

most_similar()

test_WEAT()

test_RND()

Examples

tab_similarity(demodata, cc("king, queen, man, woman"))
tab_similarity(demodata, cc("king, queen, man, woman"),
               unique=TRUE)

tab_similarity(demodata, cc("Beijing, China, Tokyo, Japan"))
tab_similarity(demodata, cc("Beijing, China, Tokyo, Japan"),
               unique=TRUE)

## only n1 * n2 word pairs across two sets of words
tab_similarity(demodata,
               words1=cc("king, queen, King, Queen"),
               words2=cc("man, woman"))

PsychWordVec documentation built on Aug. 21, 2025, 5:53 p.m.