Honours and Masters project

This project focuses on modeling heart rate data captured via FitBit Sense devices worn by team members in collaborative situations such as supervision meetings, group teaching, or nursing simulation scenarios. The primary goal is to identify stressful situations or similar events by analysing heart rate variations.

This project focuses on the automated classification of teachers' activities and co-teaching behaviors using positioning data captured via sensors and microphone data. The main task involves developing and applying machine learning techniques to analyse multimodal datasets, combining positioning and speech data to identify and categorize various teaching activities. By leveraging large language models (LLMs), Generative AI (GenAI), and Natural Language Processing (NLP), the project aims to extract features that enhance the accuracy and effectiveness of these classification tasks.

I am seeking students doing Honours or a minor thesis in a Masters interested in working on designing Learning Analytics innovations to study classroom proxemics by analysing and visualising indoor positioning data (along with other sources of data such as audio, physiological activity and characteristics of the students).

Are you ready to dive into the future of education and revolutionise how software projects are assessed? Join this innovative project aimed at creating cutting-edge learning analytics capabilities within the Faculty of IT at Monash University. This project seeks to provide automated support for teaching staff in augmenting the marking of software development and design assignments, specifically software projects submitted to the FIT-based GIT lab platform, using advanced large language models (LLMs).

This project aims to enhance student engagement and comprehension by combining Data Comics with Generative AI. Data Comics present complex information in an engaging, accessible format, and by leveraging AI, we seek to automate their creation, making the process efficient and scalable. This project involves a human-centred design approach with students and teachers to ensure the content is relevant and pedagogically sound. The collaboration will tailor Data Comics to meet the needs of learners, while AI will enable the rapid generation of personalized educational materials.

As the number of mental health patients increases, the demand for qualified counselors is on the rise. However, training/practice sessions with actual patients are often limited, let alone meeting a sufficient number of patients of different personalities. This project aims to use large language models to simulate therapy sessions under certain predefined circumstances. This project is co-supervised by a collaborator from the Psychology department in Jeffrey Cheah School of Medicine and Health Sciences.

A recommender system is a subclass of information filtering/retrieval system that provides suggestions for items that are most pertinent to a particular user without an explicit query. Recommender systems have become particularly useful in this information overload era and have played an essential role in many industries including Medical/Health, E-Commerce, Retail, Media, Banking, Telecom and Utilities (e.g., Amazon, Netflix, Spotify, Linkedin etc).

Chatbots for mental health are shown to be helpful for preventing mental health issues and improving the wellbeing of individuals, and to ease the burden on health, community and school systems.  However, the current chatbots in this area cannot interact naturally with humans and the types of interactions are limited to short text, predefined buttons etc. In contrast, psychologists in real-world interact with patients with multiple modalities, including accustic and visual information. Non-textual information is also essential for health observation and treatments of patients.

Machine learning (ML) training and evaluation usually involve large-scale datasets and complicated computation. To process data efficiently, a promising solution is to outsource the processes to cloud platforms. However, traditional approaches of collecting users' data at cloud platforms are vulnerable to data breaches. Specifically, during the ML model training or inference service offering, the cloud server could learn the input data used to train the model, model structures, user queries and inference results, which may be sensitive to users or companies. 

This project is based on the paper "Academic Journals, Incentives, and the Quality of Peer Review: A Model", in which we analyse strategic interactions between scientists and science journals.  Our results shed light on how different objectives for journals shape the strategies that scientists adopt when aiming to publish their work. In this project, we aim to extend this model to include the influence of different environmental factors such as prestige, affiliations or career stage of the scientists.