Annual Workshop on Language Technologies and NLP for Africa’s Poorly Endowed Languages - NTeALan

Annual Workshop on Language Technologies and NLP for Africa’s Poorly Endowed Languages

Workshop theme

Advances and challenges of NLP for African languages : how far have we come ?

As part of the activities of its general assembly marking the end of the year 2020 and the beginning of 2021, the association New Technologies for African Languages organizes, for the second time, a scientific workshop around the issues of NLP with regards to poorly endowed languages of Africa. This edition will take place on February 19, 2021 simultaneously online on NTeALan’s webinar platform and at the association’s headquarters (NTeALan multimedia center), located in Douala-Makepe, Tradex Parcours vita (opposite “Majestic Pressing”).

This workshop will bring together NTeALan’s large community of members living in Cameroon and abroad, invited researchers, partners and other members of the scientific community. This workshop aims at bringing together professionals, researchers and experts in African languages and Natural Language Processing (NLP), whose research work focuses on machine learning techniques and electronic lexicography applied to NLP and language pedagogy/didactics. We will also discuss the main challenges that arise in this context for the constitution of corpora in African languages. We will define possible directions for future progress.

In summary, this workshop is an opportunity for the participants and the audience to :

  • review outstanding work on the constitution of datasets useful for the  African languages processing.
  • present the latest advances in NLP applied to African languages.
  • present the platforms, datasets and softwares currently available on the market for processing African languages.
  • evaluate and challenge the NLP approaches, methods and techniques currently used to process African languages.
  • boost digital industries favourable to the digitisation of African languages.

 

To build their presentations for the workshop, speakers will be able todraw inspiration from the following themes (non-exhaustive list) :

  • Electronic Lexicography
  • Codification and Xmlisation of dictionaries
  • Knowledge Representation
  • Methods for building up lexical resources
  • Tokenization, Lemmatization, POS
  • Application of lexicographical data for NLP tools
  • Machine translation for African languages
  • Speech recognition and synthesis in African languages
  • Semantic classification of documents
  • Speech synthesis / Audio recognition
  • Named-entities Recognition in African languages
  • Question-answer system, Chatbot, NLU
  • Text extraction from images, PDFs, etc.

 

Scientific committee:

  • Pr Jules Assoumou
  • Pr Ndibnu Messina
  • Dr Ornella Wandji
  • M. Elvis MBONING

 

Contacts : levismboning@ntealan.org ; workshop@ntealan.org

Workshop day
Start time
February 19, 2021
09:00:00

Main Speakers

Pr Jules ASSOUMOU

9h15 – 9h45 UTC+1

Head of Department of African Language Linguistics at the University of Douala

Developed theme

The linguistic question in the emergence of Africa

(1) Pr Sonja Bosch & Dr Gertrud Faaß

9h45 – 10h15 UTC+1

(10h45 – 11h15 UTC+1, Johnnesburg time)

(1) Professor in the Department of African Languages at the University of South Africa (UNISA)

Developed theme

A Learners’ Dictionary for Zulu

MCF. Ndibnu Messina

10h15 – 10h45 UTC+1

Researcher at ENS and University of Yaoundé I

Developed theme

The lexicon of African minority languages and the construction of corpora in NLP

(1) Dr Gertrud Faaß & Pr Sonja Bosch

10h45 – 11h15 UTC+1

(10h45 – 11h15 UTC+1, Berlin time)

(1) Lecturer at the University of Hildesheim

Developed theme

Working towards a Zulu verb valence lexicon

M. Elvis MBONING

11h30 – 12h00 UTC+1

President of the NTeALan association

Developed theme

NTeALan collaborative dictionaries: what are the advantages of  natural language processing for African languages ?

Zouleiha

Mme Zouleiha Alhadji

12h00 – 12h30 UTC+1

PHD student at the University of Ngaoundere

Thème développé

A suffix-stripping Algorithm And transducers for the peul language

paul dayang

(1) Dr Paul Dayang &
(2) M. Jules Paulin Bayang Souloukna

12h30 – 13h00 UTC+1

(1) Senior lecturer at the University of Ngaoundere & (2) PHD student at the University of Ngaoundere

Thème développé

Computing perplexity values for under-resourced languages using ngram and deep learning approaches

Dr Rodrigue Tchamna

13h00 – 13h30 UTC+1

(7h00 – 7h30 UTC-5, New York time)

Research Associate at City College of New York and member of Resulam Association

Developed theme

The linguistic question in Presentation of Resulam’s Works, Resurrection of ancestral mother tongues, for the revitalization of African languages