ADEFSA Natural Language Processing (NLP) Bootcamp

The ADEFSA Natural Language Processing (NLP) Bootcamp was run by IFRA from the 8th to the 11th of November, as part of a programme to support the development of French higher education in Africa (ADESFA) financed by the French Ministry of Europe and Foreign Affairs (MEAE). ADEFSA has supported the University Paris Nanterre to help the University of Ibadan to set up a Masters Degree in Natural Language Processing.

This project aims to increase the power and quality of existing Nigerian national education, through joint teaching and the implementation of new pedagogical practices. This project will eventually lead to setting up a University of Ibadan Master's degree. This degree will prepare students from the humanities as well as the exact sciences to be operational in the automatic information processing and linguistic engineering sector. The skills acquired belong to linguistics, statistics and computer sciences. The target audience is students who are not yet specialized in ICST, with diverse academic backgrounds (humanities, linguistics, statistics and computer science, etc.) and who are destined for various careers (communication, business, police and security, higher education and research, etc.).

It is also worth mentioning that students in the humanities are predominantly female and computer science students are male. This type of gateway allows female students who did not think they were destined to become engineers to do so.

Finally, Nigeria is a linguistic mosaic with more than 500 languages listed. The training of Nigerian specialists in NLP is therefore a major issue at the societal and heritage level: constitution of computerized resources for the different languages, NLP tools for the development of regional languages (especially Yoruba, Igbo, Hausa and Nigerian Pidgin), automatic translation software between Nigerian languages and the languages of the international community (English, French, etc.).

The NLP Boot camp was Co-organised by Prof. Akin Odebunmi (Director, Centre for Excellence in Teaching and Learning) and Prof. Ayodele Jegede (Director, Research Management Office).

It consisted in a presentation of the curriculum of the future NLP Masters diploma to an audience of over 33 candidates from the whole of Nigeria who had applied to attend this presentation and be introduced to the various aspects of NLP. Each course of the Masters diploma will be taught in Ibadan by a team composed of (minimum) one Nigerian lecturer or expert trained and supervised by a French colleague. The French colleagues share with their Nigerian counterpart their notes and teaching material, bibliography etc. and help them prepare the course.
Courses to be covered by the future NLP programme include: General Introduction to NLP; Corpus Linguistics and processing tools; Text statistics; Parallel corpora and machine translation; Speech Recognition, Analysis and Synthesis; Databases for linguistics; Structured documents; Syntactic treebanks and parsing; Text mining and information retrieval; Distributional semantics and vector spaces; Machine translation.

Non-academic experts came from the private sector to share their experience in NLP: Dr Adegbola on Adapting NLP techniques for African languages ; David Adelani on the Challenges and Opportunities of NLP for African languages; Tolutope Ogunremi on State-of-the-Art Methods of Speech Recognition; Koko Godswill on NLP Practical Guides using NLTK.

The participants all expressed their eagerness to take part in this curriculum and hoped the University of Ibadan will soon include it in the diplomas it offers at the Master level. Prof. Akin Odebunmi assured us of his commitment to have the Master’s curriculum approved by UI authorities within the next few months.