10 Jan 2023

Call for Low Resource Natural Language Processing Postdoctoral fellowship applications

Data Science for Social Impact (DSFSI) Research group calls for interested parties


The Data Science for Social Impact (DSFSI) research group and ABSA Chair of Data Science in the Department of Computer Science, is calling for qualified persons interested to apply for a Postdoctoral Research Fellowship. The Postdoctoral Research Fellow will be required to focus on the following:

  • Research connected to the topic area. (see below)
  • Participate in broader research activities in the DSFSI Lab
  • Co-supervision of students

Topic Area: Code-switching and mixing classification modelling for low-resource South African languages.

Code-switching and code-mixing are when bilingual speakers switch between multiple languages during written or spoken communication . Both are complex processes that involve different levels of code-switching or code-mixing in accordance with proficiency in the languages used . The current language models which use natural language processing and speech models break down when people use different codes/languages for different contexts or interrupt the language processing pipeline (the systems are built only for one language), forcing users to communicate in unnatural manners. Four significant challenges occur in building code-switching and code-mixing models:

  • A shortage of code-switched data on which to train language models;
  • An absence of multilingual language annotations on code-switched standards which are needed to train supervised models;
  • Little understanding of how to leverage monolingual and parallel resources to build better-code-switching models;
  • and ultimately, how to employ these models to learn and generate new concepts when code-switching happens across multiple languages

Remuneration and Duration

  • The expected remuneration is R450K per annum.
  • The current duration is for 1 year but can be extended if more funding is secured.
  • We hope for the successful candidate to start as soon as possible (ideally by end of Q1 2023).

Closing date for applications: 29 January 2023

Apply here https://bit.ly/DSFSIPostdoc

Note: Do share with those who might be interested in this opportunity. If there are any questions,, feel free to get in contact with Prof. Vukosi Marivate (vukosi.marivate@cs.up.ac.za)

When you send an email, please use this subject.

Subject: LRL NLP PostDoc Opportunity opportunity.