Projects

Student Projects

2023

You can view the project presentations online [Youtube Playlist]

  • SABC News Corpus Classification using Machine Learning and Transformer. Abiola Akinbowale
  • Identifying South African Infrastructure-related News Articles using CorEx Topic Modelling Adinda Behrens
  • Standing on the shoulders of giants! Exploring the effectiveness of pre-trained language models on unseen low-resource languages Bayanda Kutshwa
  • Exploring Patterns from the South African Police Service Tweets Mabel Mapuwei
  • Topic Modelling of South African Government Speeches with LDA, NMF and BERTopic Murendi Rampai
  • Exploring Public Sentiment and Themes in the Post Office Dataset Panashe Mabwe
  • Extracting Atoms for the TshiVenda Language. Pandelani Nekhumbe
  • Developing a Natural Language Processing-driven categorisation system for SABC news corpus Refiloe Kekana
  • ANALYSING SOUTH AFRICAN POLICE TWEETS USING TOPIC MODELING TECHNIQUES Reshay Mahepal
  • Unsupervised Sentiment Analysis: A Deep Dive into Understanding Twitter Users Sentiment Towards the South African Post Office Sachin Modi
  • Bathong! Exploring South African slang in Twitter Discourse Thokozani Msezane
  • 2022

    You can view the project presentations online [Youtube Playlist]

  • Assessing Impact of Tokenizing or Cleaning Emojis on a Sentiment Analysis Task for a Low Resource Language Johann Pienaar
  • Fine-grained multi-label classification of sexism using Natural Language Processing Techniques Fiskani Banda
  • Twitter Sentiment Analysis on JSE Related Tweets Tsepang Polaki
  • Multi-label Classification of Rap Lyrics Matthew Grobler
  • Word representation techniques on neural networks Matthew Dickson
  • South African mines host community sentiment analysis using a twitter API. Matome Ledwaba
  • Sentiment Analysis in Bitcoin Lister Kom
  • Impact of word embedding techniques on the classification of fake news Sylvester Mothapo
  • Emotionally driven fake news in South Africa Marc Gagiano
  • Social media sentiment analysis and topic modelling for the South African insurance companies Penelope Matloga
  • E-Commerce Text Classification Tyler Ramnath
  • Detection of Fake Online Job Posts Nandi Mnguni
  • Sentiment analysis regarding access to free healthcare by immigrants in South Africa Godwin Sichulu
  • Investigating Sentiment Analysis Approaches for Low-Resource Languages Victor Dippenaar
  • A Comparison of Popular Word Embedding Techniques for Hyperpartisan News Detection with a small dataset Stephan Kersop
  • Fake news classification: exploring misclassification by topic Arno Preez
  • Aspect-based Sentiment Analysis: Aspect Limiting and Performance Kris Hamersma
  • Topical Hate Speech Detection in Multilingual Datasets using State-of-the-Art Transformer Language Models Trishanta Srikissoon
  • The Ukriane War according to Twitter - GSDMM Topic extraction on Tweets regarding the war Neacail Hilhorst
  • Fake News Detection in a South African Context - using modern NLP Transformers Joshua Da Silva
  • Use of NLP Techniques to assess life insurance claims Xolile OReilly
  • Sentiment Analysis on an African Language Kapei Sebesho

2021

You can view the project presentations online [Youtube Playlist]

  • Evaluating NLP techniques for extracting job ads on Twitter Ndamulelo Nemakhavhani
  • Unsupervised cross-lingual word embedding representation for English-isiZulu Derwin Ngomane
  • Classification for Setswana News Matimba Shingange
  • Semi-supervised learning approaches for predicting South African political sentiment for the local government elections Mashadi Ledwaba
  • COVID-19 Economic Impact Analysis: A Topic Modelling Approach Erika Scholtz
  • A comparison of classification algorithms for identifying fake news Katlego Mongoai
  • Comparison of medicinal use on covid-19 recoveries using machine learning Ntsika Faku
  • An ensemble classification of Covid-19 Fake News from Social Media Nelson Phora
  • Interpretable ML on Fake News detection Tinashe Chinyati
  • Exploring different visualisations for comparing NLP approaches Michiel van Staden
  • Trilogy Topic Modelling applied to Aylien COVID-19 dataset Shaun Johnson
  • Efficient Transformers for Hate Speech Detection Jesse Parvess
  • Online Abuse analysis with Election Twitter Data Christiaan Klopper

2020

  • Classification of Hate Speech and offensive language on twitter texts Andani Madodonga
  • Fake News Detection: Performance Analysis of Word to Vectorizing Approaches in Conjunction with Machine Learning Approaches Abhay Lal
  • Investigating text properties of fake news misclassified by state-of-the-art model BERT Hermann Ulrich Tapamo
  • Diversity in South African COVID-19 News Isak Daniel Ungerer
  • Unreliable News Detection using Machine Learning Mohlatlego Nakeng
  • Hyperpartisan News Detection using Transfer Learning Njabulo Mncwabe
  • Exploratory analysis of COVID-19 government tweets for time series approach on topic modeling in South Africa (SA) Nozuko Twala
  • Analysing the Effect of Text Augmentation Techniques on a South African Fake News Text Classification Task Pieter van der Walt
  • Fake news detection Sakhiwo Mtwenka
  • Lockdown Sentiment Analysis Semaka Malapane
  • Public Sentiment Evolution and Infection Spread during Covid-19 in South Africa – A Social Media Study Shalom Kgofelo
  • Fake News Detection Using Classification Models – a Model Comparison Study Siyabonga Mbonambi
  • Comparative Study on BERT, LTSM and Naive Bayes in Predicting Fake News Yolanda Nkalashe

2019

  • Fake News Detection using Machine Learning Techniques Abel Mnyakeni
  • A well-read machine? Towards a supervised NLP Transfer Learning Architecture Andreas Bayer
  • Using transfer learning to predict disinformation Carel Nchachi
  • Explainable hyper-partisan news detection using NLP: an investigation Frederik Pretorius
  • Investigating Attention based CNN for Fake news detection Henk Kotze
  • Stance Detection Using Natural Language Processing Hlompho Lekaka
  • On Fake News Detection: Stance detection Khutso Sepuru
  • Can NLP detection of disinformation in company annual reports be an indicator of risk? Lamont Theron
  • The Experimental Evaluation of Machine Learning Classifiers to Perform News Reliability Verification Lehan Lötter
  • To Identify Optimal Multi-Class Machine Learning Algorithm for Hate Speech Detection on Twitter Linae Munsamy
  • Topic analysis and feature optimisation to predict the likelihood of disinformation Michelle Terblanche
  • Reduce Bias In Hate Speech Detection Palesa Lepamo
  • Deceptive Opinions - Fake Amazon Reviews Retshidisitswe Lehata
  • Hate Speech Classification and Training Extension to Emails Riaal Domingues
  • That’s fake news!! Semaka Malapane