A Model for Semi-Supervised Keyword Extraction from Word Embedding Models

A fast and computationally efficient algorithm designed to enable researchers to efficiently and quickly extract semantically-related keywords using a fitted embedding model. For more details about the methods applied, see Chester (2025). .

About

An R package designed to enable researchers to quickly and efficiently generate customized sets of keywords.

For more information about the methods underlying this package see Chester (2025).

Installation

Install the package from CRAN:

install.packages("keyclust")

Or install the development version from GitHub:

install.packages("devtools") # If not already installed

devtools::install_github("pchest/keyclust")

Usage

Creating a cosimilarity matrix from a pre-fitted word embeddings model

library(keyclust)

simmat <- wordemb_FasttextEng_sample |>
    process_embed(words = "words") |>
    similarity_matrix(words = "words")

Extracting a semantically-related set of keywords from a cosimilarity matrix

seed_months <- c("october", "november")

out_months <- keyclust(simmat, seed_words = seed_months, max_n = 10)

## Initializing with october, november 
## Added september 
## Added august 
## Added february 
## Added january 
## Added december 
## Added march 
## Added april 
## Added june

out_months |>
  terms() |>
  head(n = 10)

##         Term Group_similarity
## 1    october        0.9764174
## 2   february        0.9727986
## 3  september        0.9715995
## 4    january        0.9695413
## 5   november        0.9693111
## 6     august        0.9677877
## 7      march        0.9673854
## 8      april        0.9653796
## 9       june        0.9625061
## 10  december        0.9609312