QUT offers a diverse range of student topics for Honours, Masters and PhD study. Search to find a topic that interests you or propose your own research topic to a prospective QUT supervisor. You may also ask a prospective supervisor to help you identify or refine a research topic.
Found 661 matching student topics
Displaying 25–36 of 661 results
Transport big data analytics: Imputing missing data
The missing data problem is often unavoidable for real-world data collection systems because of a variety of factors, such as sensor malfunctioning, maintenance work, transmission errors, and so on. Filling in missing information in a dataset is an important requirement for many machine-learning algorithms that require a complete dataset as input. Data imputation algorithms aim at filling the missing information in a dataset. Many missing data imputation techniques exist in the literature, with applications demonstrated on various types of datasets. …
- Study level
- PhD, Master of Philosophy, Honours
- Faculty
- Faculty of Engineering
- School
- School of Civil and Environmental Engineering
- Research centre(s)
- Centre for Data Science
Efficient parameter estimation for agent-based models of tumour growth
Cancer is an extremely heterogeneous disease, particularly at the cellular level. Cells within a single cancerous tumour undergo vastly different rates of proliferation based on their location and specific genetic mutations. Capturing this stochasticity in cell behaviour and its effect on tumour growth is challenging with a deterministic system, e.g. ordinary differential equations, however, is possible with an agent-based model (ABM). In an ABM, cells are modelled as individual agents that have a probability of proliferation and movement in each …
- Study level
- Master of Philosophy, Honours
- Faculty
- Faculty of Science
- School
- School of Mathematical Sciences
- Research centre(s)
- Centre for Data Science
Making the most of many models
In the age of Big Data, machine learning methods, and modern statistics the adage "all models are wrong but some are useful" has never been so true. This project will investigate data science approaches where more than one model makes sense for the data. Is it better to choose a single model or is there something to be gained from multiple models?This project will look at variable selection methods, penalised regression, Bayesian model averaging and conformal prediction. The research has …
- Study level
- Honours
- Faculty
- Faculty of Science
- School
- School of Mathematical Sciences
- Research centre(s)
- Centre for Data Science
Multi-modal sentiment analysis
In deep learning models, language models and word embedding methods have become popular to understand the context of text data. Popular language models such as BERT have limitations in terms of the token length. There exist some corpora that have longer text with an average of 1000 tokens. Additionally, these corpora are text-heavy and only include some images.In our prior works, we have developed several multi-modality models on social media datasets.
- Study level
- Master of Philosophy, Honours
- Faculty
- Faculty of Science
- School
- School of Computer Science
- Research centre(s)
- Centre for Data Science
Overcoming the challenges of sensitive data via synthetic data generation (case study)
In the 21st Century, there is an abundance of data, often containing insights that could benefit a number of stakeholders. However, despite this opportunity, it is often the case that the data is sensitive and can not be released by organisations or government agencies due to privacy concerns. One possible solution to the above dilemma is to instead carefully construct a 'twin' data set that contains similar information (and ideally, the same insights) as the original data set, but without …
- Study level
- Honours
- Faculty
- Faculty of Science
- School
- School of Mathematical Sciences
- Research centre(s)
- Centre for Data Science
Using time-controlled grazing to sequester carbon in Queensland rangelands
Time-controlled grazing (TCG), or cell grazing is a management strategy in which cattle are stocked and rotated across small paddocks or “cells” according to fodder availability. Grazing takes place in short durations at high stocking densities, in an effort to mimic the grazing patterns of wild ungulate herds.This management strategy has gained traction in recent years due to claims that it improves both pasture productivity and diversity, whilst also increasing long-term carbon pools. Limited data is available on the impact …
- Study level
- PhD, Master of Philosophy, Honours
- Faculty
- Faculty of Science
- School
- School of Biology and Environmental Science
- Research centre(s)
- Centre for Agriculture and the Bioeconomy
Capturing the impact of patient variability in a novel cancer treatment
In 2015, the Food and Drug Association (FDA) approved a lab-engineered virus for the treatment of melanoma (skin cancer). Since then, there has been a significant increase in the number of lab-grown viruses that are being tested in clinical trials as potential treatments of cancer. Unfortunately, it seems that a large number of patients in these clinical trials fail under this treatment and currently there is no way to distinguish between responders and non-responders to treatment.Fortunately, we can use mathematics …
- Study level
- Honours
- Faculty
- Faculty of Science
- School
- School of Mathematical Sciences
- Research centre(s)
- Centre for Data Science
Increasing resilience of robotic systems through quickest change detection technology
Future robotics systems are likely to benefit from having an ability to self-diagnose self-failure or the presence of anomalous situations (so that they can switch to fallback or fail-safe modes). Example situations include subtle sensor or actuator failure and cyber security or physical intruder detection.Such low signal-to-noise anomaly detection or self-diagnose problems can be understood using powerful mathematical and statistical tools which QCR has a rich history of advancing through collaboration with industry partners and publication in premium international venues.
- Study level
- PhD
- Faculty
- Faculty of Engineering
- School
- School of Electrical Engineering and Robotics
Process-data governance patterns
Data is recognised a strategic asset for organisations. There is a growing need to manage the voluminous data an organisation is exposed to in order to use it for decision-making.Of particular significance is process data, which consists of information about the execution of processes. Such information is used to uncover behaviour of processes within an organisation. This brings forth the significance of data governance. Data governance is the exercise of control and authority over management of data. Despite its significance, …
- Study level
- PhD, Master of Philosophy, Honours
- Faculty
- Faculty of Science
- School
- School of Information Systems
Unified def-site and use-site security policies for component-based software systems
Securing the information manipulated by computer systems, such as privacy and integrity in social software, is a challenge. Traditional methods to impose limits on the information disclosure, such as access control lists, firewalls, and cryptography, provide no guarantees about information propagation. For instance, cryptography provides no guarantees about the confidentiality of the data are given once it is decrypted.Information flow control (IFC) is the problem of ensuring secure information flow according to specified policies within computer systems. Modern applications are …
- Study level
- PhD, Master of Philosophy
- Faculty
- Faculty of Science
- School
- School of Computer Science
Productive reproducible workflows for deep learning-enabled large-scale industry systems
Deep learning is a mainstream to increase the capability of industry systems, particularly for those with massive data input and output. It is seen that many tools are now claimed to be freely available and could facilitate such process of development and deployment significantly with scalability and quality.However, limited attention has been on developing reproducible and productive workflows to identify the tools and their values towards large-scale industry systems. In this project, we will explore how to design such a …
- Study level
- PhD, Master of Philosophy
- Faculty
- Faculty of Science
- School
- School of Computer Science
Systematic evaluation towards the analysis of open-source supply chain on ML4SE tasks
Applying machine learning algorithms to source code related SE task is rapidly developing and attracts the attention from both researchers and industry engineers. While there are many program languages available, applying such techniques, i.e., the representation learning models, for different languages may achieve different performance. Particularly, they all have their own strict syntax, which determines the abstract syntax tree. Thus, a lot of different open-source supply chain are available, for example the parsing tools are used to build AST from …
- Study level
- PhD, Master of Philosophy
- Faculty
- Faculty of Science
- School
- School of Computer Science
Contact us
If you have questions about the best options for you, the application process, your research topic, finding a supervisor or anything else, get in touch with us today.