predict.word2vec: Predict functionalities for a word2vec model

In word2vec: Distributed Representations of Words

Predict functionalities for a word2vec model

Description

Get either

• the embedding of words

• the nearest words which are similar to either a word or a word vector

Usage

## S3 method for class 'word2vec'
predict(
  object,
  newdata,
  type = c("nearest", "embedding"),
  top_n = 10L,
  encoding = "UTF-8",
  ...
)

Arguments

object	a word2vec model as returned by word2vec or read.word2vec
newdata	for type 'embedding', newdata should be a character vector of words for type 'nearest', newdata should be a character vector of words or a matrix in the embedding space
type	either 'embedding' or 'nearest'. Defaults to 'nearest'.
top_n	show only the top n nearest neighbours. Defaults to 10.
encoding	set the encoding of the text elements to the specified encoding. Defaults to 'UTF-8'.
...	not used

Value

depending on the type, you get a different result back:

• for type nearest: a list of data.frames with columns term, similarity and rank indicating with words which are closest to the provided newdata words or word vectors. If newdata is just one vector instead of a matrix, it returns a data.frame

• for type embedding: a matrix of word vectors of the words provided in newdata

See Also

word2vec, read.word2vec

Examples

path  <- system.file(package = "word2vec", "models", "example.bin")
model <- read.word2vec(path)
emb <- predict(model, c("bus", "toilet", "unknownword"), type = "embedding")
emb
nn  <- predict(model, c("bus", "toilet"), type = "nearest", top_n = 5)
nn

# Do some calculations with the vectors and find similar terms to these
emb <- as.matrix(model)
vector <- emb["buurt", ] - emb["rustige", ] + emb["restaurants", ]
predict(model, vector, type = "nearest", top_n = 10)

vector <- emb["gastvrouw", ] - emb["gastvrij", ]
predict(model, vector, type = "nearest", top_n = 5)

vectors <- emb[c("gastheer", "gastvrouw"), ]
vectors <- rbind(vectors, avg = colMeans(vectors))
predict(model, vectors, type = "nearest", top_n = 10)

word2vec documentation built on Oct. 8, 2023, 1:07 a.m.