CarbPred

Using machine learning to predict nuclear magnetic resonance spectra for carbohydrates via graph neural networks

By: Channing Bellamy, Unays Bhad, Matthew Dean

Supervised By: Michelle Kuttel, Jan Buys

Carbohydrates are complex molecules, critical to drug and vaccine development. Their study often involves the analysis of nuclear magnetic resonance (NMR) spectra, which are challenging to interpret, even for experts. We build upon prior work, GeqShift,...

CS Honours CS Honours Project

Toward Robust Malware Classification

Data Augmentation, Topology Optimisation, and Concept Drift Monitoring

By: Shaylin Velen, Callum Musselwhite, Simphile Mkhize

Supervised By: Geoff Nitschke

Our project investigates the application of various machine learning techniques for malware detection. As malware engineers improve and release new variants, identifying whether a file is malicious becomes increasingly challenging. This malware evolution...

CS Honours CS Honours Project

Measuring the effects of climate control on presence in natural virtual environments

VRClimate

By: Pietro Scheepers, Zahra Bawa, Rachel Strachan

Supervised By: James Gain

Virtual Reality, or VR, is defined as a technology that allows a user to interact with a computer-simulated environment, whether that environment is a simulation of the real world or an imaginary world. Currently, the main focus of creating these virtual...

CS Honours CS Honours Project

SOILAI

Generating Soilscapes with Random Forets

By: Josh Britz, James Ridley

Supervised By: James Gain, Patrick Marais

Soils support plant life and form an important aspect of ecological biodiversity. Soil data, however, is often absent from digital terrains but nevertheless remain important in areas of computational ecology. Digital elevation maps are a common artefact...

CS Honours CS Honours Project

Evolutionary Methods in Collective Herding Systems

By: Emily Raine, Naheed Khan

Supervised By: Geoff Nitschke

This project extended a 2D sheep-herding multi-agent system, where dog agents herd sheep into a target zone. We investigated the effectiveness of hybrid evolutionary methods in dynamic environments and compared the performance of agents with fixed versus...

CS Honours CS Honours Project

LLMASP: Synthesising Answer Set Programs with LLMs

By: Rowan Rosenberg, Julyan Van Der Westhuizen, Keegan O'Brien

Supervised By: Jesse Heyninck

Answer Set Programming (ASP) is a declarative programming framework. It is a powerful tool for solving complex logical and combinatorial problems, with many modern, real-world applications. However, its power is limited by a phenomenon known as the...

CS Honours CS Honours Project

EXTRC

By: Julia Cotterrell, Nevaniah Gounden, Jethro Dunn

Supervised By: Tommie Meyer

This project explores how defeasible reasoning, a form of non-monotonic logic, can be extended and made more interpretable beyond Rational Closure (RC). The optimisation of RC was combined with natural language processing for knowledge base (KB) creation...

CS Honours CS Honours Project

Improving Research Question Quality with Controlled Natural Languages and Large Language Models

RQ-CNL

By: Rector Ratsaka, Emma Van Der Berg, Mandikudza Dangwa

Supervised By: Zola Mahlaza

Formulating a well-defined research question (RQ) is a crucial yet often challenging step in the research process, particularly for trainee researchers. Poorly constructed RQs can lead to ambiguous objectives and ineffective methodologies, reducing the...

CS Honours CS Honours Project

BabyLMs for Sample-Efficient Language Modelling of Ancient Greek

Investigating the suitability of BabyLM architectures for language modelling in the low-resource context of Ancient Greek.

By: Dimitri Dalakas, Stelio Dalakas, Kimon Christelis

Supervised By: Francois Meyer

Pretrained Language Models (PLMs) have achieved state-of-the-art performance across a wide range of Natural Language Understanding tasks in recent years. This success has relied on vast amounts of training data, which are not readily available for...

CS Honours CS Honours Project

Dungeons and Developers

Evaluating Real-Time Competitive Game-based Learning in Introductory Computer Science

By: Mahir Moodaley, Kai Connock, Ibrahim Abdou

Supervised By: Gary Stewart

Many first-year students, especially those with no previous experience, have difficulties learning and understanding Computer Science. This is due to a variety of factors including lack of engagement and low motivation. Students who have difficulty with...

CS Honours CS Honours Project

Securing the digital highway

Investigating the performance of fuzzers on Wireguard-Go

By: Thalia Hawthorn, Michal Sacks

Supervised By: Josiah Chavula

Network-based tools have become a central component in the operation of many institutions such as businesses, universities and governments. The growing amount of sensitive data being transported over networks necessitates innovation to ensure these...

CS Honours CS Honours Project

2025 Computer Science Honors Generative AI Applied to Student Guide (GAISG) Project

GAISG

By: Angelo Yang, Nova Adams-Duma, Daanyaal Ballim

Supervised By: Gary Stewart

The Generative AI Applied to Student Guide (GAISG) project reimagines UCT’s “Science is Tough: But So Are You” student guide as an interactive, multimodal learning platform. Traditional PDF-based guides often fail to engage students, limiting...

CS Honours CS Honours Project

POL-ID

AI-Based South African Honey Authentication

By: Yash Ramklass, Maryam Mather, Lillian Mtumanje

Supervised By: Patrick Marais, Cesarina Edmonds-Smith, Janais Delport

Authenticating the botanical origin of honey is crucial for ensuring product integrity, consumer trust, and sustainable trade. However, manual pollen identification under a microscope is time-consuming and prone to human error. To address this, our...

CS Honours CS Honours Project

3DHAR-AI

Human Activity Recognition and Prediction using Deep Neural Networks

By: Kishalan Pather, Andrew Erasmus, Temi Aina

Supervised By: Deshen Moodley

The growing field of HAR sees many applications from remote health monitoring for the elderly to smart homes that adapt to our daily routines. In our work, we investigated the performance of three state-of-the-art spatio-temporal graph neural networks...

CS Honours CS Honours Project

AI Share Trader

By: Kabelo Mbayi, Siyanda Makhathini, Saul Chipwayambokoma

Supervised By: Deshen Moodley

Traditional deep learning models for stock forecasting often overlook the complex interdependencies between financial assets. This study investigates the efficacy of Spatio-Temporal Graph Neural Networks (ST-GNNs) for stock prediction on the Johannesburg...

CS Honours CS Honours Project

deepEEG

Deep Learning for Infant-Brain EEG Modelling

By: Tziyona Cohen, Caleb Bessit

Supervised By: Francois Meyer, Jonathan Shock

Electroencephalography (EEG) measures electrical brain activity and is a common tool for monitoring neurodevelopment and diagnosing brain disorders. However, analysing EEG data is challenging due to its complex interpretability and inter-subject...

CS Honours CS Honours Project

Fine-Tuning Large Language Models for Low-Resource Languages

By: Mitchell Johnson, Dineo Chiloane, Corey Webb

Supervised By: Jan Buys

Most Large Language Models (LLMs) have been trained either on English text only or on text from multiple languages, but with a severe imbalance between data from different languages due to orders of magnitude differences in available web text per...

CS Honours CS Honours Project

IsiXhosa Medical Machine Translation

By: Elijah Sherman, Nick Matzopoulos, Malibongwe Makhonza

Supervised By: Francois Meyer

Machine translation (MT) systems for low-resource languages like isiXhosa are still a big challenge because there aren’t many parallel datasets available for training models. Current MT systems work well when translating between isiXhosa and English for...

CS Honours CS Honours Project

READ: Developing a South African ART with Fine-tuned Mispronunciation Detection, Voice Detection, Multi-speaker Detection and Text Generation

By: Daniel Holgate, Christian Slier, Kirsten Sutherland

Supervised By: Jan Buys

Reading literacy is poor for most South Africans due to the numerous negative aftereffects of Apartheid. Both Phala et al. and Naidoo et al. refer to the Progress in International Reading Literacy Study (PIRLS) 2011, which revealed that 61% of school...

CS Honours CS Honours Project

STASP

Sports Timetabling with Answer Set Programming

By: Mahir Ahammed, Roanda Muedi

Supervised By: Jesse Heyninck

This project explores the use of Answer Set Programming (ASP) as a declarative paradigm to solve the classic combinatorial sports timetabling problem. ASP allows scheduling rules and constraints to be expressed in a high-level, human-readable form, which...

CS Honours CS Honours Project

Accurate Modelling and Effective Visualization of Carbohydrate Conformation

By: Andre Toerien, Aidan Ravenscroft

Supervised By: Michelle Kuttel, Nicholas Yerolemou

Carbohydrates play critical roles in biological processes, and their interactions depend on their 3D shapes, or conformations. Understanding these conformations, through computer modelling and visualization, is therefore key to modern drug and vaccine...

CS Honours CS Honours Project

Lightweight AI Intrusion Detection

LAIDS

By: Sian Caine, Claire Campbell, Christopher Blignaut

Supervised By: Josiah Chavula

Intrusion detection systems (IDS) monitor network traffic for malicious activity. However, many AI-based IDSs are too resource-heavy for edge deployment, leaving resource-constrained portions of networks unguarded. The main objective of this study are to...

CS Honours CS Honours Project

Investigating Deep Learning Solutions to Fish Identification and Classification

By: Joseph Goldblatt, Catalina Althoff-Thomson, Max Elkington

Supervised By: Patrick Marais, Jonathan Shock

Fishery managers, marine biologists, ecologists, and conservationists all rely on accurate estimates of fish population size and composition. Recently, there has been a growing interest in the use of Remote Underwater Video (RUV) methods for fish...

CS Honours CS Honours Project

Traffic Flow

An Analysis of STGNNs through Visualization

By: Raaziq Parkar, Mohammad Arshad Hussain

Supervised By: Deshen Moodley

Traffic is a highly dynamic system containing spatial, temporal and random patterns. Spatial Temporal Graph Neural Networks(STGNNs) model traffic as an evolving graph, dynamically capturing both spatial and temporal patterns. Current evaluation metrics...

CS Honours CS Honours Project

The Dynamic Vehicle Routing Problem

By: Jayden Moore, Ryan Schapiro, Aidan Brand

Supervised By: Krupa Prag

The Dynamic Vehicle Routing Problem (DVRP) extends the classical Vehicle Routing Problem by incorporating dynamic and uncertain elements, such as stochastic customer requests, making it more applicable to real-world logistics. Unlike static routing where...

CS Honours CS Honours Project

Synthetic Data Generation for Small Data Orthopedic Radiology

By: Max Van Veen, Oliver Foxcroft, Ethan Topat

Supervised By: Geoff Nitschke, Bilal Aslan

Address the data scarcity issues faced by automated medical classification methods using machine learning models for synthetic data generation with an appropriate set of metrics to evaluate their quality and diversity performance.​

CS Honours CS Honours Project

A Performance and Resilience Analysis of SANReN

A Spectral and Centrality Metric Framework for Resilient Network Topology Design and Analysis

By: Kerry-Lynn Whyte, Nicolas Wise, Alex White

Supervised By: Josiah Chavula, Michelle Kuttel

The South African National Research and Education Network (SANReN) provides the high-speed connectivity required for research collaboration and education but SANReN faces challenges related to rising demand, congestion, and vulnerability to node and edge...

CS Honours CS Honours Project

IsiPython: A Localised IsiXhosa Programming Environment for Inclusive Computing Education

Combining Language, Learning, and Code to Make Programming More Accessible

By: Lutho Mngqibisa, Buhle Mlandu

Supervised By: Aslam Safla

The dominance of English in programming languages creates barriers for South African students whose primary language is not English, particularly in early Computer Science education. This forces learners to understand programming concepts while...

CS Honours CS Honours Project

Data Driven Control

Reinforcement Learning versus Classical control for Control Systems

By: Ben Ruijsch van Dugteren, Nathan Wells, Nicholas Cristaudo

Supervised By: Krupa Prag

This study examines the application of deep reinforcement learning to data-driven control and benchmarks it against classical proportional–integral–derivative control. Two system classes are considered: mechanical systems (inverted pendulum and...

CS Honours CS Honours Project

Redirected Walking in VR

By: Prashanth Padiachy, Oscar Field, Kyle Fagan

Supervised By: James Gain

Our project explores how different VR locomotion techniques can be combined with Redirected Walking (RDW) to enable natural interaction with physical props in small tracked spaces. We developed three systems in Unity using the HTC Vive Pro: Free Teleport...

CS Honours CS Honours Project

MyAdvisor

Virtual Curriculum Assistant

By: Dylan Howard, Ethan Allies, Mitchell Flanegan

Supervised By: Aslam Safla, Nicholas Yerolemou

MyAdvisor is a centralized hub for UCT advising, built to simplify handbooks, streamline advisor access, and support degree planning. Instead of navigating scattered portals and static PDFs, students get one clear, interactive system that works on both...

CS Honours Project