Package: conText 2.0.0

conText: 'a la Carte' on Text (ConText) Embedding Regression

A fast, flexible and transparent framework to estimate context-specific word and short document embeddings using the 'a la carte' embeddings approach developed by Khodak et al. (2018) <arxiv:1805.05388> and evaluate hypotheses about covariate effects on embeddings using the regression framework developed by Rodriguez et al. (2021)<https://github.com/prodriguezsosa/EmbeddingRegression>.

Authors:Pedro L. Rodriguez [aut, cre, cph], Arthur Spirling [aut], Brandon Stewart [aut], Christopher Barrie [ctb]

conText_2.0.0.tar.gz
conText_2.0.0.zip(r-4.5)conText_2.0.0.zip(r-4.4)conText_2.0.0.zip(r-4.3)
conText_2.0.0.tgz(r-4.4-any)conText_2.0.0.tgz(r-4.3-any)
conText_2.0.0.tar.gz(r-4.5-noble)conText_2.0.0.tar.gz(r-4.4-noble)
conText_2.0.0.tgz(r-4.4-emscripten)conText_2.0.0.tgz(r-4.3-emscripten)
conText.pdf |conText.html✨
conText/json (API)

# Install 'conText' in R:

install.packages('conText', repos = c('https://prodriguezsosa.r-universe.dev', 'https://cloud.r-project.org'))

Bug tracker:https://github.com/prodriguezsosa/context/issues

Datasets:

cr_glove_subset - GloVe subset
cr_sample_corpus - Congressional Record sample corpus
cr_transform - Transformation matrix

On CRAN:

9.65 score 97 stars 1.7k scripts 226 downloads 29 exports 58 dependencies

Last updated 7 months agofrom:c373ea228e. Checks:OK: 1 NOTE: 6. Indexed: yes.

Target	Result	Date
Doc / Vignettes	OK	Oct 31 2024
R-4.5-win	NOTE	Oct 31 2024
R-4.5-linux	NOTE	Oct 31 2024
R-4.4-win	NOTE	Oct 31 2024
R-4.4-mac	NOTE	Oct 31 2024
R-4.3-win	NOTE	Oct 31 2024
R-4.3-mac	NOTE	Oct 31 2024

Exports:bootstrap_contrast bootstrap_nns compute_contrast compute_transform conText contrast_nns cos_sim dem dem_group dem_sample embed_target feature_sim fem find_nns get_context get_cos_sim get_grouped_similarity get_local_vocab get_ncs get_nns get_nns_ratio get_seq_cos_sim ncs nns nns_ratio permute_contrast plot_nns_ratio prototypical_context tokens_context

Dependencies:cli colorspace cpp11 data.table digest dplyr fansi farver fastDummies fastmatch float generics ggplot2 glue gtable isoband ISOcodes jsonlite labeling lattice lgr lifecycle magrittr MASS Matrix MatrixExtra mgcv mlapi munsell nlme pillar pkgconfig plyr purrr quanteda R6 RColorBrewer Rcpp RcppArmadillo reshape2 RhpcBLASctl rlang rsparse scales SnowballC stopwords stringi stringr text2vec tibble tidyr tidyselect utf8 vctrs viridisLite withr xml2 yaml

Finite Sample Bias

Rendered fromfinite_sample_bias.Rmdusingknitr::rmarkdownon Oct 31 2024.

Last update: 2023-08-04
Started: 2023-08-04

Quick Start Guide

Rendered fromquickstart.Rmdusingknitr::rmarkdownon Oct 31 2024.

Last update: 2023-08-04
Started: 2021-03-06

Help page	Topics
Bootstrap similarity and ratio computations	bootstrap_contrast
Bootstrap nearest neighbors	bootstrap_nns
Bootstrap OLS	bootstrap_ols
Boostrap similarity vector	bootstrap_similarity
build a 'conText-class' object	build_conText
build a 'dem-class' object	build_dem
build a 'fem-class' object	build_fem
Compute similarity and similarity ratios	compute_contrast
Compute similarity vector (sub-function of bootstrap_similarity)	compute_similarity
Compute transformation matrix A	compute_transform
Embedding regression	conText
Contrast nearest neighbors	contrast_nns
Compute the cosine similarity between one or more ALC embeddings and a set of features.	cos_sim
GloVe subset	cr_glove_subset
Congressional Record sample corpus	cr_sample_corpus
Transformation matrix	cr_transform
Build a document-embedding matrix	dem
Average document-embeddings in a dem by a grouping variable	dem_group
Randomly sample documents from a dem	dem_sample
Embed target using either: (a) a la carte OR (b) simple (untransformed) averaging of context embeddings	embed_target
Given two feature-embedding-matrices, compute "parallel" cosine similarities between overlapping features.	feature_sim
Create an feature-embedding matrix	fem
Find cosine similarities between target and candidate words	find_cos_sim
Return nearest neighbors based on cosine similarity	find_nns
Get context words (words within a symmetric window around the target word/phrase) sorrounding a user defined target.	get_context
Given a tokenized corpus, compute the cosine similarities of the resulting ALC embeddings and a defined set of features.	get_cos_sim
Get averaged similarity scores between target word(s) and one or two vectors of candidate words.	get_grouped_similarity
Identify words common to a collection of texts and a set of pretrained embeddings.	get_local_vocab
Given a set of tokenized contexts, find the top N nearest contexts.	get_ncs
Given a tokenized corpus and a set of candidate neighbors, find the top N nearest neighbors.	get_nns
Given a corpus and a binary grouping variable, computes the ratio of cosine similarities over the union of their respective N nearest neighbors.	get_nns_ratio
Calculate cosine similarities between target word and candidates words over sequenced variable using ALC embedding approach	get_seq_cos_sim
Given a set of embeddings and a set of tokenized contexts, find the top N nearest contexts.	ncs
Given a set of embeddings and a set of candidate neighbors, find the top N nearest neighbors.	nns
Computes the ratio of cosine similarities for two embeddings over the union of their respective top N nearest neighbors.	nns_ratio
Permute similarity and ratio computations	permute_contrast
Permute OLS	permute_ols
Plot output of 'get_nns_ratio()'	plot_nns_ratio
Find most "prototypical" contexts.	prototypical_context
Run jackknife debiased OLS	run_jack_ols
Run OLS	run_ols
Get the tokens of contexts sorrounding user defined patterns	tokens_context

Package: conText 2.0.0

conText: 'a la Carte' on Text (ConText) Embedding Regression

Finite Sample Bias

Quick Start Guide

Citation

Development and contributors

Readme and manuals

Help Manual

Usage by other packages (reverse dependencies)