maltese-gender-debias

Gender bias detection using distributional models within the Maltese language.

https://github.com/andaqu/maltese-gender-debias

Science Score: 44.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • CITATION.cff file
    Found CITATION.cff file
  • codemeta.json file
    Found codemeta.json file
  • .zenodo.json file
    Found .zenodo.json file
  • DOI references
  • Academic publication links
  • Academic email domains
  • Institutional organization owner
  • JOSS paper metadata
  • Scientific vocabulary similarity
    Low similarity (5.0%) to scientific vocabulary
Last synced: 9 months ago · JSON representation ·

Repository

Gender bias detection using distributional models within the Maltese language.

Basic Info
  • Host: GitHub
  • Owner: andaqu
  • Language: Jupyter Notebook
  • Default Branch: main
  • Homepage:
  • Size: 472 KB
Statistics
  • Stars: 0
  • Watchers: 1
  • Forks: 0
  • Open Issues: 0
  • Releases: 0
Created over 5 years ago · Last pushed over 2 years ago
Metadata Files
Readme Citation

README.md

Project structure:

gender-debias ├── report.pdf ├── dataset (*) │ ├── dataset-neutral │ ├── dataset-original │ └── dataset-raw ├── saved │ ├── neutral (*) │ ├── original (*) │ └── data └── src ├── debiaswe ├── utils └── model.ipynb

(*) not included in repository. Please refer to Google Drive link [1].

  • report.pdf: Coursework report.
  • dataset-neutral: Lemmatised pre-processed dataset.
  • dataset-original: Pre-processed dataset.
  • dataset-raw: Original obtained dataset.
  • saved/neutral: Models trained from lemmatised pre-processed dataset (word2vec and fasttext).
  • saved/original: Models trained from original pre-processed dataset (word2vec, fasttext and tf-idf vectorizer).
  • saved/data: Saved data for gender bias observation and mitigation.
  • src/debiaswe: Obtained and arranged debiaswe code from [2]. (Note to self :: A better approach would have been to fork this project and have it as a submodule...)
  • src/utils: Utility scripts to pre-process and other tasks.
  • model.ipynb: Jupyter notebook to train and test models.

[1]: https://bit.ly/3krABOw
[2]: https://github.com/tolga-b/debiaswe

Owner

  • Name: Andrew Aquilina
  • Login: andaqu
  • Kind: user

Citation (CITATION.cff)

cff-version: 1.2.0
authors:
- family-names: "Andrew Aquilina"
title: "maltese-gender-debias"
version: 1.0.0
date-released: 2021-02-26
url: "https://github.com/andaqu/maltese-gender-debias"

GitHub Events

Total
Last Year