MIT 808: Big Data Science Capstone Project

Welcome to the University of Pretoria MIT808: Big Data Science Capstone Project online exhibition

2025 Exhibition

The 2025 Capstone Exhibition showcases 9 impactful student projects that tackle urgent and diverse challenges using cutting-edge Data Science techniques. This year, 18 students worked in interdisciplinary teams and with real-world partners to create solutions in conservation, public health, climate science, political discourse, and law.

👉 You can view all 2025 projects under the Projects tab or explore the organizations and researchers we collaborated with via the Partners tab.

---

🚀 2025 Project Overviews

🐜 Matabele Ant Raids: Optimising Termite Control

Focus: Predicting raid effectiveness using XGBoost and behavioral data from the Mpala Research Centre.
Impact: A simulation tool for ecologists and farmers to model termite control strategies.

🗳️ Political Polarisation in SA Election Discourse

Focus: Analysing xenophobia-related political discourse on Twitter using DistilRoBERTa.
Impact: Mapping polarisation trends and their evolution across elections since 2016.

🌧️ Deep Learning for Rainfall Forecasting

Focus: CNN-LSTM and fine-tuned GraphCast models for medium-range (3–10 day) rainfall prediction in Southern Africa.
Impact: A deployed forecasting tool for weather services and disaster preparedness teams.

🐘 ElephantMatcher: Wildlife Population Monitoring

Focus: Object detection and feature comparison to identify elephants from aerial and ground images.
Impact: Automated pipeline for conservationists to track and re-identify elephants.

🐘 Elephant Identification from Camera Footage

Focus: Using ResNet-50 and clustering to distinguish individual elephants from videos.
Impact: Semi-supervised model with a user feedback loop for iterative model refinement.

🎶 Decoding Music Law with NLP

Focus: LegalBERT, logistic regression, and random forest models applied to public submissions on music law reform.
Impact: A tool for legal insight extraction to empower creatives and inform policymakers.

🐄 Predictive Dashboard for Foot and Mouth Disease (FMD)

Focus: Integrating environmental and livestock data to model FMD outbreak risks.
Impact: A dashboard to support early warning systems and livestock health monitoring.

🧠 Zoonotic Disease Awareness in Sepedi

Focus: Fine-tuned GPT-4o and Gemini LLMs to deliver health education about rabies in Sepedi.
Impact: A culturally relevant chatbot that improves public health communication in low-resource languages.

📊 FMD Outbreak Trends in the Beef Value Chain

Focus: Creating an integrated data platform combining climate, livestock, and historical outbreak data.
Impact: Interactive visualisations and predictive modelling for informed disease management planning.

---

Prior Years

View projects from

• 2025 capstone
• 2024 capstone
• 2023 capstone
• 2022 capstone
• 2021 capstone
• 2020 capstone

MIT 808 Information

The module, from 2020, has been taught by Vukosi Marivate and Abiodun Modupe. Vukosi Marivate has a background in Machine Learning and Artificial Intelligence and is interested in the role of Data Science in Society.Abiodun Modupe is interested in the confluence of artificial intelligence for natural language processing and speech recognition (because of the abundance of text documents and the need for knowledge to be extracted). In this module, students carry out a Data Science Capstone project that brings together the theoretical module they have completed in the first year of their MITC Big Data Science program.

More information: - 📘 MIT808 Public Website - 🧑‍🎓 MIT Big Data Science Programme - 📬 Mailing List for Updates

---

✉️ Feedback and Contact

We’d love to hear from you!
📩 dsfsi.info@up.ac.za