GV

GeezerVison

The World Wide Web, restored to 800 pixels of glory.

Home|About|Gaming cabinet|Archives|How it works|Starter sites

On this site

  • Sign in
  • dsfsi
  • ActionsCI/CD and automation
  • CodespacesCloud development environments
  • All Features
  • Documentation
  • Training
  • Certifications
  • Build with Copilot
  • Customize Copilot
  • Integrations
  • Trending repos
  • GitHub Sponsors
  • Developer articles

This view

  • Original page
  • New address

GeezerVison is showing github.com — Data Science for Social Impact Research Group @ University of Pretoria

README.md

We are the Data Science for Social Impact research group at the Computer Science Department, University of Pretoria.

Our general areas of work straddle Data Science for Society as well as Local Language Natural Language Processing. These two strands are complementary. Our work in Data Science and Society has allowed us to have a more nuanced approach to understanding the systematic challenges that face being able to do excellent science with local languages. Through Data Science for Society, we have to understand how when one carries through Data Science research, we situate how the users are part of the process. We find that we need to adjust our research to take care of these challenges and innovate in ways we gather direct data or alternative data.

For us, Data Science for Society means being able to improve approaches/methods or scientific tools for DS while enhancing the ways decision-makers can use the insights that come from these tools. Local Language Natural Language Processing is focused on ways to develop new tools, new data and methodology to improve the state of African languages.

DSFSI Vision, Mission and Values.

Vision

To be a leading inclusive lab that creates and harnesses data and multidisciplinary scientific exploration for societal impact.

Mission

Data-driven collaborative innovation to empower society to tackle challenges and preserve our languages.

Values

  • Community and Collaboration
  • Shared responsibility
  • Inclusiveness
  • Integrity and openness
  • Agency
  • Generosity

Pinned Loading

  1. vukuzenzele-nlp vukuzenzele-nlp Public

    Forked from dsfsi/dsfsi-dataset-template

    The dataset contains editions from the South African government magazine Vuk'uzenzele. Data was scraped from PDFs that have been placed in the data/raw folder. The PDFS were obtained from the Vuk'u…

    Jupyter Notebook 8 6

  2. textaugment textaugment Public

    TextAugment: Text Augmentation Library

    Python 447 62

  3. covid19za covid19za Public

    Coronavirus COVID-19 (2019-nCoV) Data Repository and Dashboard for South Africa

    Jupyter Notebook 259 198

  4. dsfsi-datasets dsfsi-datasets Public

    Official DSFSI Public Datasets Registry - Comprehensive catalog of 50+ datasets for South African & African languages. Includes speech recognition, NLP, terminology, health, legal & financial data…

    Jupyter Notebook 12 8

  5. deadlines deadlines Public

    Forked from vukosim/ai-ds-africa-deadlines

    ⏰ AI/ML/DS conference/workshop/event deadlines on the African continent

    HTML 26 12

  6. za-mafoko za-mafoko Public

    DSFSI South African Terminlogy Lists and Lexicon Project

    HTML 6 1

Repositories

Loading
Type
Language
Sort
Showing 10 of 64 repositories
  • lacunafund-datasets Public Forked from Fair-Forward/datasets

    A catalog that links to the datasets and use-cases of the Lacuna Fund

    dsfsi/lacunafund-datasets's past year of commit activity
    Python 1 CC0-1.0 4 0 0 Updated Nov 12, 2026
  • lacunafund-mirror Public
    dsfsi/lacunafund-mirror's past year of commit activity
    TypeScript 0 1 0 0 Updated Nov 11, 2026
  • isbis-up-2027 Public
    dsfsi/isbis-up-2027's past year of commit activity
    CSS 0 0 0 0 Updated Nov 6, 2026
  • maswo-south-african-sign-language-dataset Public

    This repository contains code and dataset used to create the Maswao (meaning signs in Northern Sotho) dataset.

    dsfsi/maswo-south-african-sign-language-dataset's past year of commit activity
    Python 0 0 0 0 Updated Nov 1, 2026
  • vukuzenzele-nlp Public Forked from dsfsi/dsfsi-dataset-template

    The dataset contains editions from the South African government magazine Vuk'uzenzele. Data was scraped from PDFs that have been placed in the data/raw folder. The PDFS were obtained from the Vuk'uzenzele website.

    dsfsi/vukuzenzele-nlp's past year of commit activity
    Jupyter Notebook 8 MIT 7 3 0 Updated Oct 25, 2026
  • On-mathematical-and-computational-models-for-low-resource-languages. Public

    On mathematical and computational models for low-resource languages.

    dsfsi/On-mathematical-and-computational-models-for-low-resource-languages.'s past year of commit activity
    0 0 0 0 Updated Oct 13, 2026
  • deadlines Public Forked from vukosim/ai-ds-africa-deadlines

    ⏰ AI/ML/DS conference/workshop/event deadlines on the African continent

    dsfsi/deadlines's past year of commit activity
    HTML 26 1,080 0 1 Updated Jul 18, 2026
  • cross-lingual-transfer-gains-evaluation Public

    Code and datasets for studying cross-lingual embedding transfer and transfer gain analysis in mutually intelligible African languages using VecMap and MUSE.

    dsfsi/cross-lingual-transfer-gains-evaluation's past year of commit activity
    Python 0 0 0 0 Updated Jun 15, 2026
  • dsfsi-podcast Public

    Uh oh!

    There was an error while loading. Please reload this page.

    dsfsi/dsfsi-podcast's past year of commit activity
    0 0 0 0 Updated Apr 17, 2026
  • up-manuscript Public

    A template for MSc, MIT and PhD that meets UP requirements.

    Uh oh!

    There was an error while loading. Please reload this page.

    dsfsi/up-manuscript's past year of commit activity
    TeX 0 GPL-3.0 0 0 0 Updated Mar 11, 2026
View all repositories

People

@vukosim

Top languages

Python Jupyter Notebook HTML SCSS Makefile

Most used topics

dsfsi-datasets south-africa natural-language-processing dataset nlp
You are visitor 01402503 — GeezerVison does not store the pages you look up.
Best viewed with Netscape Navigator 4.7 or Internet Explorer 6 at 800×600. No cookies required.