Find Nearest Neighbors Based on Cosine Similarity

This function identifies the nearest neighbors of a given term or vector in a matrix based on cosine similarity.

sem_nearest_neighbors(x, matrix, n = 10)

Arguments

x

A character or numeric vector representing the term or vector.

matrix

A numeric matrix or a sparse matrix against which the similarity is calculated.

n

Number of nearest neighbors to return.

Value

A data frame with the ranks, terms, and their cosine similarity scores.

Examples

if (FALSE) {
 api_url <- "https://api-inference.huggingface.co/pipeline/feature-extraction/sentence-transformers/all-MiniLM-L6-v2"
 api_token <- ''
 matrix <- api_huggingface_embeddings(tif,
                                      text_hierarchy = c('doc_id', 'sentence_id'),
                                      api_token = api_token,
                                      api_url = '')
 query <- api_huggingface_embeddings(query = "Where's the party at?",
                                     api_token = api_token,
                                     api_url = api_url)
 neighbors <- sem_nearest_neighbors(x = query, matrix = matrix)
}