pairwise_similarity: Cosine similarity of pairs of items

In dgrtwo/widyr: Widen, Process, then Re-Tidy Data

Cosine similarity of pairs of items

Description

Compute cosine similarity of all pairs of items in a tidy table.

Usage

pairwise_similarity(tbl, item, feature, value, ...)

pairwise_similarity_(tbl, item, feature, value, ...)

Arguments

tbl	Table
item	Item to compare; will end up in item1 and item2 columns
feature	Column describing the feature that links one item to others
value	Value
...	Extra arguments passed on to squarely(), such as diag and upper

See Also

squarely()

Examples

library(janeaustenr)
library(dplyr)
library(tidytext)

# Comparing Jane Austen novels
austen_words <- austen_books() %>%
  unnest_tokens(word, text) %>%
  anti_join(stop_words, by = "word") %>%
  count(book, word) %>%
  ungroup()

# closest books to each other
closest <- austen_words %>%
  pairwise_similarity(book, word, n) %>%
  arrange(desc(similarity))

closest

closest %>%
  filter(item1 == "Emma")

dgrtwo/widyr documentation built on Nov. 14, 2022, 4:07 a.m.