Filter by faculty:

Found 14 matching student topics

Displaying 1–12 of 14 results

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, Vacation research experience scheme
Faculty
Faculty of Science
School
School of Computer Science
Research centre(s)
Centre for Data Science

Evaluation of language models and word embedding methods for natural language processing applications

In deep learning models, language models and word embedding methods have become popular to understand the context of text data. There exist many variants of these methods and have different limitations. This project will introduce you to the hot topic of language models and the fields of Natural Language Processing and Text Mining. 

Study level
Honours, Vacation research experience scheme
Faculty
Faculty of Science
School
School of Computer Science
Research centre(s)
Centre for Data Science

Building explainable and trustworthy intelligent systems

Existing machine learning-based intelligent systems are autonomous and opaque (often considered “black-box” systems), which has led to the lack of trust in AI adoption and, consequently, the gap between machine and human being.In 2018, the European Parliament adopted the General Data Protection Regulation (GDPR), which introduces a right of explanation for all human individuals to obtain “meaningful explanations of the logic involved” when a decision is made by automated systems. To this end, it is a compliance that an intelligent …

Study level
PhD, Master of Philosophy, Honours, Vacation research experience scheme
Faculty
Faculty of Science
School
School of Information Systems
Research centre(s)
Centre for Data Science

Explainable AI-enabled predictive analytics

Modern predictive analytics underpinned by AI-enabled learning (such as machine learning, deep learning) techniques has become a key enabler to the automation of data-driven decision making. In the context of process monitoring and forecast, predictive analytics has been applied to making predictions about the future state of a running process instance - for example, which task will be carried out next, when and who will perform the task, when will an ongoing process instance complete, what will be the outcome …

Study level
PhD, Master of Philosophy, Honours, Vacation research experience scheme
Faculty
Faculty of Science
School
School of Information Systems
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, Vacation research experience scheme
Faculty
Faculty of Science
School
School of Mathematical Sciences
Research centre(s)
Centre for Data Science

Predicting failure in robotic vision

Computer vision models predict where objects are in an image, and what those objects are. In robotics, these vision models are used to allow robots to perceive their environment and choose safe and smart actions based on this perception.Computer vision models can fail silently when exposed to unexpected or difficult environments - e.g. changes in camera viewpoints, changes in lighting, or when seeing new objects that haven't been seen before. This raises concerns about the safety of using vision models …

Study level
Honours, Vacation research experience scheme
Faculty
Faculty of Engineering
School
School of Electrical Engineering and Robotics
Research centre(s)
Centre for Robotics

Fine-grained software vulnerability detection using deep learning techniques

Software vulnerability is a major threat to the security of software systems. Thus, the successful prediction of security vulnerability is one of the most effective attack mitigation solutions. Existing approaches for software vulnerability detection (SVD) can be classified into static and dynamic methods. Powered by AI capabilities, especially with the advancement of machine learning techniques, current software has been produced with more sophisticated methodologies and components. This has made the automatic vulnerability proneness prediction even more challenging. Recent research efforts …

Study level
PhD, Master of Philosophy
Faculty
Faculty of Science
School
School of Computer Science

Deep Learning for engineering object recognition

Study level
Vacation research experience scheme
Faculty
Faculty of Engineering
School
School of Mechanical, Medical and Process Engineering

Deep learning for robotics in open-world conditions

To fully integrate deep learning into robotics, it's important that deep learning systems can reliably estimate the uncertainty in their predictions. This allows robots to treat a deep neural network like any other sensor and use the established Bayesian techniques to fuse the network’s predictions with prior knowledge or other sensor measurements or to accumulate information over time.Deep learning systems typically return scores from their softmax layers that are proportional to the system’s confidence. They are not calibrated probabilities and …

Study level
PhD, Master of Philosophy, Honours, Vacation research experience scheme
Faculty
Faculty of Engineering
School
School of Electrical Engineering and Robotics

Deep learning for robotics in open-world conditions: Uncertainty, continuous learning, active learning

In order to fully integrate deep learning into robotics, it is important that deep learning systems can reliably estimate the uncertainty in their predictions. This would allow robots to treat a deep neural network like any other sensor, and use the established Bayesian techniques to fuse the network’s predictions with prior knowledge or other sensor measurements or to accumulate information over time.Deep learning systems, e.g. for classification or detection, typically return scores from their softmax layers that are proportional to …

Study level
PhD
Faculty
Faculty of Engineering
School
School of Electrical Engineering and Robotics

A physics-guided deep learning-based framework for computational mechanics

Computational mechanics is an essential discipline that uses numerical schemes to approximately solve mechanics problems. It provides engineers with precious knowledge about the structures to identify the at-risk area and further guide the structural design and optimisation process.Deep learning (DL) is an important branch of machine learning (ML). The great success of the DL techniques has been witnessed in the past decade. Now, various fields have benefited from the DL techniques, including computer vision, financial prediction, and bioinformatics. Therefore, it …

Study level
Vacation research experience scheme
Faculty
Faculty of Engineering
School
School of Mechanical, Medical and Process Engineering

Data reasoning to extend domain knowledge in deep learning

A wide variety of companies now use personalized prediction models to improve customer satisfaction, for example, detecting cancer relapses, Detecting Attacks in Networks (e.g., SDN) or understanding Customer Online Shopping Behaviour. However, the dramatic increase in size and complexity of newly generated data from various sources is creating a number of challenges for domain experts to make personalized prediction.For example, early detection of cancer can drastically improve the chance and successful treatment. Recently, supervised deep learning has brought breakthroughs in …

Study level
PhD, Master of Philosophy, Honours, Vacation research experience scheme
Faculty
Faculty of Science
School
School of Computer Science
Research centre(s)
Centre for Data Science

Page 1 of 2