Honours and Masters project

 Title: Health and Social Challenges of Refugee Populations in Australia: A Data-Driven Investigation.

Keywords: Refugees, health outcomes, social challenges, data integration, policy analysis

Project Description: Perform a comprehensive data-driven study on health and social challenges faced by refugee populations in Australia. The project integrates multiple datasets, applies advanced statistical analysis, and uses machine learning to detect patterns that inform policy and support services.

This project uses machine learning and predictive analytics to group customers based on their shopping habits using publicly available or synthetic transactional datasets. Students will clean and analyse purchase data, apply clustering algorithms such as K-Means and Hierarchical Clustering, and identify common product purchase patterns using association rule mining. The project aims to show how data-driven methods can help businesses better understand customer behaviour and design targeted marketing strategies.

 Title: Grouping Customers by Shopping Habits with Machine Learning

Specialised project:

This project applies machine learning and predictive analytics to detect early signs of heart disease using publicly available cardiovascular datasets. Students will clean and analyse health data, apply algorithms such as Decision Trees and Random Forest, and identify key risk factors for heart disease. The project aims to show how data-driven methods can support early intervention and improve patient outcomes.

Title: Early Detection of Heart Disease Using Machine Learning and Predictive Analytics

This research aims to bridge a critical accessibility gap in digital navigation tools by developing an inclusive, intelligent system that combines map services, street-level imagery, and large language models (LLMs). Current systems often fail to support marginalised users—such as older adults, people with vision impairments, or those with limited mobility—by overlooking nuanced environmental cues such as footpath obstructions, ramp availability, or visibility of building entrances. By democratising navigation, the project addresses both a technological and equity gap in urban mobility.

Urbanisation and climate change are accelerating environmental degradation, making cities critical battlegrounds for sustainability.

In human-AI collaboration, it is essential for AI systems to understand and anticipate human behavior in order to coordinate effectively. Conversely, humans also form inferences about the agent’s beliefs and goals to facilitate smoother collaboration. As a result, AI agents should adapt their behavior to align with human reasoning patterns, making their actions more interpretable and predictable. This principle forms the foundation of transparent planning (MacNally et al, 2018).

Since the COVID-19 pandemic there has been an increasing shift within higher education away from traditional lectures and towards asynchronous content delivery through pre-recorded videos. This has a number of benefits: students can consume content at their own pace, videos can be reused, and production value can be increased. However, academics typically have no training or experience in video production, so pre-recorded videos are most often just a simulacrum of a standard lecture (i.e., a slideshow with voiceover).

Access to education is an important issue. A major factor preventing access can be the cost of textbooks, which is a significant barrier for some students. Open Education Resources (OERs) are a popular option for reducing this financial burden, as they are free to any person with an internet connection. 

Introductory programming remains a significant challenge for many students. A large factor impacting success is each student's motivation to engage with assessment and practice exercises. One strategy for improving student engagement is to offer multiple assessment options.

Leveraging the FLoRA adaptive learning platform, we will conduct a five-phase research program combining experimental studies and advanced trace data analysis. Through time-stamped interaction data, we aim to detect behavioural signals of metacognitive disengagement using machine learning and time-series modeling techniques. These insights will inform the development of adaptive scaffolding tools that encourage students to monitor, evaluate, and adjust their learning strategies when using GenAI.