conText - GitLite

KO | EN

gitlite — search

#python #hacktoberfest #ai #typescript #android #machine-learning #rust #deep-learning #java #awesome #kubernetes #vue #pytorch #mcp

conText

(★ 111)

An R package for estimating and doing statistical inference on context-specific word embeddings.

파일 탐색기

최종 버전 다운로드 (.zip)

- - R
  - Rscript
- 00check.log
- - rhub.yaml
- .DS_Store
- cr_glove_subset.rda
- cr_sample_corpus.rda
- cr_transform.rda
- .DS_Store
- cr_glove_subset.R
- cr_sample_corpus.R
- cr_transform.R
- .DS_Store
- bootstrap_contrast.Rd
- bootstrap_nns.Rd
- bootstrap_similarity.Rd
- build_conText.Rd
- build_dem.Rd
- build_fem.Rd
- compute_contrast.Rd
- compute_similarity.Rd
- compute_transform.Rd
- conText-class.Rd
- conText-package.Rd
- conText.Rd
- contrast_nns.Rd
- cos_sim.Rd
- cr_glove_subset.Rd
- cr_sample_corpus.Rd
- cr_transform.Rd
- dem-class.Rd
- dem.Rd
- dem_group.Rd
- dem_sample.Rd
- embed_target.Rd
- feature_sim.Rd
- fem-class.Rd
- fem.Rd
- find_cos_sim.Rd
- find_nns.Rd
- get_context.Rd
- get_cos_sim.Rd
- get_grouped_similarity.Rd
- get_local_vocab.Rd
- get_ncs.Rd
- get_nns.Rd
- get_nns_ratio.Rd
- get_seq_cos_sim.Rd
- ncs.Rd
- nns.Rd
- nns_ratio.Rd
- permute_contrast.Rd
- plot_nns_ratio.Rd
- prototypical_context.Rd
- run_ols.Rd
- tokens_context.Rd
- bootstrap_nns.R
- classes.R
- compute_transform.R
- conText-package.R
- conText.R
- contrast_nns.R
- cos_sim.R
- data.R
- dem.R
- dem_group.R
- dem_sample.R
- embed_target.R
- feature_sim.R
- fem.R
- find_nns.R
- get_context.R
- get_cos_sim.R
- get_grouped_similarity.R
- get_local_vocab.R
- get_ncs.R
- get_nns.R
- get_nns_ratio.R
- get_seq_cos_sim.R
- globals.R
- ncs.R
- nns.R
- nns_ratio.R
- object-builder.R
- plot_nns_ratio.R
- prototypical_context.R
- tokens_context.R
- zzz.R
- bias_in_distance_measures.md
- bias_in_distance_measures.Rmd
- .DS_Store
- quickstart.md
- quickstart.Rmd
.gitignore
.Rbuildignore
conText.Rproj
DESCRIPTION
NAMESPACE
README.md

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

![logo-conText](https://user-images.githubusercontent.com/6556873/138291456-5dd454d2-b37c-478a-8e37-6b2b4c20710e.jpeg)

# About

**conText** provides 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)](https://arxiv.org/abs/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).

# How to Install

`install.packages("conText")`

# Datasets

To use **conText** you will need three objects:

1.  A (quanteda) **corpus** with the documents and corresponding document variables you want to evaluate.
2.  A set of (GloVe) **pre-trained embeddings**.
3.  A **transformation matrix** specific to the pre-trained embeddings.

**conText** includes sample objects for all three but keep in mind these are just meant to illustrate function implementations. In [this Dropbox folder](https://www.dropbox.com/sh/jsyrag7opfo7l7i/AAB1z7tumLuKihGu2-FDmhmKa?dl=0) we have included the raw versions of these objects including the full Stanford GloVe 300-dimensional embeddings (labeled *glove.rds*) and its corresponding transformation matrix estimated by Khodak et al. (2018) (labeled *khodakA.rds*). We provide an equivalent RDS file for the [2024 GloVe embeddings](https://nlp.stanford.edu/projects/glove/) released in July 2025 (labeled \_glove_2024.rds).

# Quick Start Guides

Check out this [Quick Start Guide](https://github.com/prodriguezsosa/conText/blob/master/vignettes/quickstart.md) to get going with `conText` (last updated: 04/17/2026).

# Latest Updates

04/17/2026: Version 3.0.1 is now available on CRAN. The update patches a bug with with the jackknife_fraction argument in the `conText()` function. Users should now see decreases in runtime for `jackknife_fraction < 1` which we recommend to use when testing code on large datasets (but users should aim to obtain standard errors using the full sample before publishing results. Additionally, see updated guidance on for running the jackknife in parallel in the [Quick Start Guide](https://github.com/prodriguezsosa/conText/blob/master/vignettes/quickstart.md).

07/28/2025: As noted in [Rodriguez et al. (2023)](https://www.cambridge.org/core/journals/american-political-science-review/article/embedding-regression-models-for-contextspecific-description-and-inference/4C90013E5C714C8483ED95CC699022FB) (p. 1272), distance measures typically used to compare representations in high-dimensional space (such as embedding vectors) exhibit statistical bias. In [Green et al. (2025)](https://www.cambridge.org/core/journals/political-analysis/article/measuring-distances-in-high-dimensional-spaces/88126F4A48F121387D249C1856C3665B), we explore the severity of this problem for text-as-data applications and provide and validate a bias correction for the squared Euclidean distance. We implement this estimator and other recommendations from the paper in the latest update to the `conText()` function. Please refer to the [Bias in Distance Measures](https://github.com/prodriguezsosa/conText/blob/master/updates/bias_in_distance_measures.md) vignette for additional information and the [Quick Start Guide](https://github.com/prodriguezsosa/conText/blob/master/vignettes/quickstart.md) for examples of how to use the new version of the function and a description of changes in the output.

# Multilanguage Resources

For those working in languages other than English, we have a set of data and code resources [here](https://alcembeddings.org/)