DSFSI @ SACAIR 2023 - Muldersdrift, South Africa: 4-8 December 2024

Data Science for Social Impact (DSFSI) is thrilled to have the opportunity to participate in and contribute to Southern African Conference on AI Research (SACAIR) 2023.

Member activities

Vukosi Marivate, PI @ DSFSI, is a panelist at the IJCAI AI for Good Panel on Tuesday and a panalist at the AI in Africa Panel on Wednesday. On Thursday Vukosi will presenting these papers:

• Vukosi Marivate, Moseli Mots’Oehli, Valencia Wagner, Richard Lastrucci, and Isheanesu Dzingirai. PuoBERTa: Training and evaluation of a curated language model for Setswana, Artificial Intelligence Research. SACAIR 2023. Communications in Computer and Information Science. 2023. [NLP] <> [Paper URL] [Preprint URL] [Dataset] [Software/Library] DOI: 10.1007/978-3-031-49002-6_17
• A. De Jager, V. Marivate, and A. Modupe. Multimodal Misinformation Detection in a South African Social Media Environment, Artificial Intelligence Research. SACAIR 2023. Communications in Computer and Information Science. 2023. [NLP][SOC] <> [Paper URL] DOI: 10.1007/978-3-031-49002-6_19

Derwin Ngomane, Masters Student @ DSFSI, attended the 2023 SACAIR unconference, delivering the opening presentation with his extended abstract titled “Predicting South African Protests using Graph Neural Networks.” In the presentation, he highlighted the benefits of employing Graph Neural Networks (GNNs) to forecast instances of protests in the South African context. Derwin’s participation in the SACAIR unconference not only provided a platform to share his research but also facilitated valuable interactions with fellow young and upcoming researchers, as well as established researchers, fostering collaboration and affirming the significance of his contributions to the field. More information of the extended abstract can be found on the conference Book of Abstracts

Rozina Myoya, Masters Student @ DSFSI,

On Friday, Rozina will present this paper

• Rozina Myoya, Lesego Matojane, Abiodun Modup, Vukosi Marivate, and Albert Myburgh. Evaluating Drone Imagery for Wildlife Unique Feature Identification, Proceedings of the Fourth Southern African Conference for Artificial Intelligence Research. 2023. [ML][SOC] <> [Paper URL]