ΑΙhub.org
 

Natural language processing model for African languages

by
17 November 2021



share this:

africa
Researchers have developed an AI model to help computers work more efficiently with a wider variety of languages.

African languages have received relatively little attention from computer scientists, so few natural language processing capabilities have been available to large swaths of the continent. A new language model, developed by researchers at the University of Waterloo’s David R. Cheriton School of Computer Science, begins to fill that gap by enabling computers to analyze text in African languages for many useful tasks.

The new neural network model, which the researchers have dubbed AfriBERTa, uses deep-learning techniques to achieve state-of-the-art results for low-resource languages.

The neural network language model works specifically with 11 African languages, such as Amharic, Hausa, and Swahili, spoken collectively by more than 400 million people. It achieves competitive output quality despite learning from just one gigabyte of text, while other models require thousands of times more data.

“Pretrained language models have transformed the way computers process and analyze textual data for tasks ranging from machine translation to question answering,” said Kelechi Ogueji, a master’s student in computer science at Waterloo. “Sadly, African languages have received little attention from the research community.”

“One of the challenges is that neural networks are bewilderingly text- and computer-intensive to build. And unlike English, which has enormous quantities of available text, most of the 7,000 or so languages spoken worldwide can be characterized as low-resource, in that there is a lack of data available to feed data-hungry neural networks.”

Most of these models work using a technique known as pretraining. To accomplish this, the researcher presents the model with text where some of the words have been covered up or masked. The model then has to guess the masked words. By repeating this process, many billions of times, the model learns the statistical associations between words.

“Being able to pretrain models that are just as accurate for certain downstream tasks, but using vastly smaller amounts of data has many advantages,” said Jimmy Lin, the Cheriton Chair in Computer Science and Ogueji’s advisor. “Needing less data to train the language model means that less computation is required and consequently lower carbon emissions associated with operating massive data centres. Smaller datasets also make data curation more practical, which is one approach to reduce the biases present in the models.”

“This work takes a small but important step to bringing natural language processing capabilities to more than 1.3 billion people on the African continent.”

Assisting Ogueji and Lin in this research is Yuxin Zhu, who recently completed an undergraduate degree in computer science at Waterloo. Together, they presented their research paper, Small data? No problem! Exploring the viability of pretrained multilingual language models for low-resource languages, at the Multilingual Representation Learning Workshop at the 2021 Conference on Empirical Methods in Natural Language Processing.



tags: ,


University of Waterloo




            AIhub is supported by:


Related posts :



The Machine Ethics Podcast: featuring Marc Steen

In this episode, Ben chats to Marc Steen about AI as tools, the ethics of business models, writing "Ethics for People Who Work in Tech", and more.
06 June 2023, by

On privacy and personalization in federated learning: a retrospective on the US/UK PETs challenge

Studying the use of differential privacy in personalized, cross-silo federated learning.
05 June 2023, by

VISION AI Open Day: Trustworthy AI

Watch the roundtable discussion on trustworthy AI, with a focus on generative models, from the AI Open Day held in Prague.
02 June 2023, by

PeSTo: an AI tool for predicting protein interactions

The model can predict the binding interfaces of proteins when they bind other proteins, nucleic acids, lipids, ions, and small molecules.
01 June 2023, by

Tetris reveals how people respond to an unfair AI algorithm

An experiment in which two people play a modified version of Tetris revealed that players who get fewer turns perceive the other player as less likeable, regardless of whether a person or an algorithm allocates the turns.
31 May 2023, by

AIhub monthly digest: May 2023 – mitigating biases, ICLR invited talks, and Eurovision fun

Welcome to our monthly digest, where you can catch up with AI research, events and news from the month past.
30 May 2023, by





©2021 - Association for the Understanding of Artificial Intelligence


 












©2021 - ROBOTS Association