Compute a matrix of cosine similarity/distance of word pairs.

Compute a matrix of cosine similarity/distance of word pairs.

Usage

pair_similarity(
  data,
  words = NULL,
  pattern = NULL,
  words1 = NULL,
  words2 = NULL,
  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 (see str_subset). If neither words nor pattern are specified (i.e., both are NULL), then all words in the data will be extracted.

words1, words2

[Option 3] Two sets of words for only n1 * n2 word pairs. See examples.

distance

Compute cosine distance instead? Defaults to FALSE (cosine similarity).

Value

A matrix of pairwise cosine similarity/distance.

Download

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

See also

cosine_similarity

plot_similarity

tab_similarity

most_similar

Examples

pair_similarity(demodata, c("China", "Chinese"))
#>             China   Chinese
#> China   1.0000000 0.7678081
#> Chinese 0.7678081 1.0000000

pair_similarity(demodata, pattern="^Chi")
#> 4 words matched...
#>             China    Chicago    Chinese      Chile
#> China   1.0000000 0.13040186 0.76780811 0.38012317
#> Chicago 0.1304019 1.00000000 0.09174141 0.08685822
#> Chinese 0.7678081 0.09174141 1.00000000 0.21538189
#> Chile   0.3801232 0.08685822 0.21538189 1.00000000

pair_similarity(demodata,
                words1=c("China", "Chinese"),
                words2=c("Japan", "Japanese"))
#>             Japan Japanese
#> China   0.5967756 0.413391
#> Chinese 0.4226447 0.642242