The workshop

The rise in ML community efforts on the African continent has led to a growing interest in Natural Language Processing, particularly for African languages which are typically low resource languages.

This interest is manifesting in the form of national, regional, continental and even global collaborative efforts to build corpora, as well as the application of the aggregated corpora to various NLP tasks.

This workshop therefore has several aims;

  1. to showcase this work being done by the African NLP community and provide a platform to share this expertise with a global audience interested in NLP techniques for low resource languages
  2. to provide a platform for the groups involved with the various projects to meet, interact, share and forge closer collaboration
  3. to provide a platform for junior researchers to present papers, solutions, and begin interacting with the wider NLP community
  4. to present an opportunity for more experienced researchers to further publicize their work and inspire younger researchers through keynotes and invited talks

Programme

 

Morning Invited & Contributed Talks

7:00-7:10Opening Remarks
  • Kathleen Siminyu
7:10-7:45Invited Talk: Masakhane, A Focus on Machine Translation for African Languages
  • Jade Abbott
7:45‑8:05Contributed Talk: Distant Supervision and Noisy Label Learning for Low Resource Named Entity Recognition: A Study on Hausa and Yorùbá
  • David Adelani
8:05 ‑8:35

AI4D Challenge Winner: Crowdsourcing a Wolof ASR dataset using a custom built tool

  • Thierno Diop

Morning Invited & Contributed Talks

8:35-8:40Poster talk: Tunizi: A Tunisian Arabizi Sentiment Analysis Dataset
  • Chayma Fourati
8:40-8:45Poster talk: The Modified Segmentation Algorithm for Recognition of Ancient Ge’ez Scripts Written on the Vellum
  • Girma Sisay
8:45‑8:50Poster talk: Unsupervised Pidgin Text Generation By Pivoting English Data and Self-Training
  • Ernie Chang
8:50‑8:55Poster talk: Using LSTM Networks to Translate from French to Senegalese Local Languages: Wolof as a Case Study
  • Alla Lo
8:55 ‑9:00Poster talk: Amharic Abstractive Text Summarization
  • Amr Zaki
9:00‑9:05

Poster talk: A Hierarchical Transformer for Unsupervised Parsing

  • Ashok Thillaisundaram
9:05‑9:10Poster talk: On optimal transformer depth for low-resource language translation
  • Elan van Biljon
9:10‑9:15

Poster talk: Combining Pretrained High-Resource Embeddings and Subword Representations for Low-Resource Languages

  • Machel Reid
9:15‑9:20Poster talk: Evaluating Amharic Machine Translation
  • Asmelash Teka Hadgu
9:20‑9:25

Poster talk: FFR V1.0: Fon-French Neural Machine Translation

  • Chris Emezue & Bonaventure Dossou
9:25‑9:30Poster talk: Using Punkt for Sentence Segmentation in non-Latin Scripts: Experiments on Kurdish (Sorani) Texts
  • Roshna Omer Abdulrahman
9:30‑9:35

Poster talk: Tigrinya Neural Machine Translation with Transfer Learning for Humanitarian Response

  • Alp Oktem
9:35‑9:40Poster talk: Invited Talk: IntrodUTCion to Multilingual Wikipedia Data for NLP
  • Isaac Johnson
9:40‑10:40LIVE Q&A with Invited & Contributed talk speakers
  • Jade Abbott, Orevaoghene Ahia, David Adelani, Theirno Diop, Isaac Johnson, moderated by Kathleen Siminyu

Evening Invited & Contributed Talks

15:00‑15:35Invited Talk: The Low-resource Natural Language Processing Toolbox, 2020 Version
  • Graham Neubig
15:35‑16:10Panel Discussion: What are African NLP’s Moonshot Problems?
  • Natalie Schluter, Sebastian Ruder, Surafel Melaku Lakew, moderated by Jade Abbott
16:10‑16:30Contributed Talk: Towards A Sign Language Gloss Representation Of Modern Standard Arabic
  • Salma El Anigri
16:30‑16:50Contributed Talk: Improving Yorùbá Diacritic Restoration
  • David Adelani
16:50‑17:00AI4D Challenge Winner: Creating MT and ASR datasets from online sources (video may not be available)
  • Dosseh Ameck Guy Max Desire

Evening Poster Session

17:00‑17:05Poster talk: Investigating an approach for low resource language dataset creation, curation and classification: Setswana and Sepedi
  • Vukosi Marivate
17:05‑17:10Poster talk: Pooled Embeddings for Unsupervised IndUTCion of Yorùbá Morphology
  • Olamilekan Wahab
17:10‑17:15Poster talk: Neural Machine Translation for South Africa’s Official Languages
  • Jason Webster
17:15‑17:20Poster talk: Assessing Human Translations from French to Bambara for Machine Learning: a Pilot Study
  • Allahsera Auguste Tapo
17:20‑17:25Poster talk: A Benchmark Dataset for Igbo-English Machine Translation
  • Ignatius Ezean
17:25‑17:30Poster talk: Bert Fine-tuning For Arabic Text Summarization
  • Khalid Elmadani
17:30‑17:35Poster talk: Semantic Enrichment of Nigerian Pidgin English for Contextual Sentiment Classification
  • Wuraola Oyewusi
17:35‑17:40Poster talk: Low Resource Neural Machine Translation: A Benchmark for Five African Languages
  • Surafel Melaku Lakew
17:40‑17:45Poster talk: Neurals Networks for Projecting Named Entities from English to Ewondo
  • Michael Franklin Mbouopda
17:45‑17:50Poster talk: Neural Machine Translation for African Languages
  • Xian Li
17:50‑17:55Poster talk: Towards Neural Machine Translation for Edoid Languages
  • Iroro Orife

Evening Q&A with Invited & Contributed talk speakers

18:00‑18:30LIVE Q&A with Invited & Contributed talk Speakers
  • Graham Neubig, Salma El Anigri, Iroro Orife, David Adelani Dosseh Ameck Guy Max Desire
18:30‑18:35Concluding remarks