School of Computing

Data Science Research Group - suggested PhD projects

Text mining using machine learning models

If you are a prospective PhD student and possess passion to process and mine interesting patterns from the text data, then please free to get in touch with Dr Shoaib Jameel. Most importantly, if you have a passion to develop novel machine learning methods through exploration of the beautiful world of Mathematics, then you are an ideal candidate! Here machine learning could mean anything related to deep learning, vector space embeddings, probabilistic topic models, among others.

Supervised by Dr Shoaib Jameel

Deep Transfer Learning for Automatic Sleep Scoring

Sleep scoring is a fundamental in sleep assessment and diagnosis and home-based sleep monitoring. Deep learning has been shown to excel on automating this task, achieving an accuracy rate comparable to manual scoring by sleep experts. However, many sleep studies suffer from the problem of insufficient data to fully utilize deep neural networks as different lab uses different recording setups, leading to the need of training automated algorithms on rather small datasets. This project is to explore transfer learning with deep neural networks to leverage large and publicly available datasets for data compensation in sleep studies with a small cohort.

Supervised by Dr Huy Phan

Machine learning for pathological speech analysis

Whereas speech and spoken language patterns may reveal a person's risk for mental disorders, such as depression and bipolar disorder. This project aims to apply machine learning/deep learning to extract acoustic and lingistic markers in the hope to identify those at risk for mental illness. Early identification and diagnosis of mental disorders could lead to better recovery.

Supervised by Dr Huy Phan

Deep learning for overlapping audio event detection

Audio event detection, which is an important task in machine hearing, is potential for many real-world applications. Deep learning has been demonstrated to work well for detection of monophonic events. However, detection of overlapping events (i.e. multiple target events happening at the same time) remains to be challenging. This project aim to explore and develop new deep network architectures for overlapping event detection. Potentially, such a network should be multitasking, i.e. being able to untangle the event mixtures and recognizing unmixed audio events at the same time.

Supervised by Dr Huy Phan

Audio scene classification with end-to-end deep learning

Recognizing an environemt via acoustic signals is foreseen for many context-aware applications. However, it is a challenging task as an audio recording could contain a lot of redundant and irrelevant information. This propject is to develop new deep learning algorithms to locate and extract critical information of interest and improve audio scene classification. Moreover, salient information extraction and classifcation should be done in the same network which is elegantly trained in an end-to-end fashtion.

Supervised by Dr Huy Phan

Predicting Recovery from Stroke with Machine Learning

This PhD research would focus on interpretable machine learning applied to data acquired from stroke patients (https://www.ucl.ac.uk/ploras/). This work will be with Professor Cathy Price (Wellcome Centre for Human Neuroimaging, UCL), whose (PLORAS) team has collected one of the largest data sets of stroke patients (greater than 1,000), including structural MRI scans, behaviour and demographics. A key focus of Cathy Price’s work is to predict the recovery trajectory of stroke patients from their structural MRI scans, particularly patients with language deficits (i.e. that are aphasic). Progress has been made on this using traditional and now deep learning methods.

Critical to clinical uptake of machine learning in this area is the ability to interpret the predictions it provides in a fashion that can be communicated to clinicians, patients and carers. The PhD student would work on this topic, using methods such as neural-symbolic techniques. The student will be located in the School of Computing at the University of Kent, but will regularly visit and work closely with Cathy Price’s team at the Wellcome Centre for Human Neuroimaging. Expertise in machine learning will be provided by Dr Thomas Hope (Wellcome Centre for Human Neuroimaging, UCL) and Dr Marek Grzes (School of Computing, University of Kent).

Supervised by Professor Howard Bowman and Marek Grzes


Machine hearing to help care for the elderly

Imagine a system that passively monitors sounds in the home of an at-risk elderly person. It doesn’t share the data but simply tries to understand it, learning their daily routine. But if something goes wrong, if there is a fall, or something smashing or some other unfortunate event, it is able to alert a carer through a text message.  The research needed to make this a reality is called machine hearing. It is a relatively new research field, and we are developing a number of systems that can help give computers the power to hear and understand the sounds around them. It is a combination of machine learning (or computational intelligence) and audio processing.

Supervised by Professor Ian McLoughlin

Bionic voice

Imagine doing research on a system that can help voice-loss patients to speak again, or to speak with a natural sounding voice for the first time in years. In this project we turn the whispery, hoarse and virtually non-existent voice of speech-loss patients into clear and natural sounding speech.  We don’t do speech synthesis, but instead take the patients own vocal signals (such as they are), analyse them and determine what they are tying to say, and then we recreate the components of their voice that are missing.

Supervised by Professor Ian McLoughlin

 

Super-audible speech

This Google funded project aims to use the movement of a persons’ mouth as a biometric, so that a smartphone can verify that the person speaking into it is really the person they claim to be. This kind of security technology is long overdue as we use our mobile devices for more and more personal and important activities. The project makes use of various signal process techniques allied with machine learning to model the mouth movements of particular users, and then detect how well the current speaker matches the model.

Supervised by Professor Ian McLoughlin

 

 

Genetic programming for temperature weather derivatives

Weather derivatives are financial instruments used as part of a risk management strategy to reduce risk associated with adverse or unexpected weather conditions. Just as traditional contingent insurance claims, whose payoffs depend upon the price of some fundamental, a weather derivative has an underlying measure, such as rainfall, temperature, humidity, or snowfall. However, in the majority of the weather derivatives, the underlying asset is a temperature index. Hence, the proposed work will be focusing on temperature weather derivatives. The problem of temperature weather derivatives can be divided into two main parts: (i) temperature prediction, and (ii) pricing of weather derivative contracts. This project will use an evolutionary approach, called Genetic Programming (GP) to predict future temperature and derive pricing equations. GP is a nature-inspired algorithm, which uses the principle of natural evolution to find computer programs that perform well in a given task. One of the main advantages of GP is its ability to perform well in high-dimensional combinatorial problems, such as the one of weather derivatives pricing.

Supervised by Dr Michael Kampouridis

 

Financial forecasting with directional changes

In the aftermath of a global financial crisis, it is more important than ever to have a better understanding of the markets and be able to forecast their movement. Directional changes (DC) is a new concept that is based on the idea that an event-based system can capture significant points in price movements that the traditional physical time methods cannot. For instance, if one was using daily closing prices, s/he would never notice the Dow Jones Industrial Average flash crash on 6 May 2010, where an almost 1000 point loss (about 9%) took place, only to recover most of those loses within minutes. Hence, instead of looking at the market from an interval-based perspective, it is proposed to record the key events in the market (e.g., changes in the stock price by a pre-specified percentage), and summarise the data based on these events.
This project will use Genetic Programming (GP) methods to create trading strategies. GP is an evolutionary technique inspired by natural evolution, where computer programs act as the individuals of a population. GP has been extensively used in the past for financial forecasting, and has shown it is able to identify patterns in financial data.

Supervised by Dr Michael Kampouridis

 

Incorporating self-evaluation into computational creativity systems

In exploring how computers can perform creative tasks, computational creativity research has produced many systems that can generate creative products or creative activity. Evaluation, a critical part of the creative process, has not been employed to such a great extent within creative systems. Recent work has concentrated on evaluating the creativity of such computational systems, but this has consisted of the system(s) being evaluated by external evaluators. Incorporation of self-evaluation into computational creativity systems would be a useful contribution to this research area. This work would also contribute towards exploring an important philosophical issue in computational creativity: is it the software that is being creative, or the programmers behind the software? In this project the PhD candidate will explore evaluation methods for evaluating both the quality of output from a creative system and the creativity of the system itself. The candidate will experiment with incorporating evaluation methods into a creative system and analyse the results to explore how computational creativity systems can incorporate self-evaluation. The creative systems studied could be in the area of musical or linguistic creativity, or in a creative area of the student's choosing.

Supervised by Dr Anna Jordanous*

 

Evaluating Computational Creativity

Computational creativity is the computational study of computer systems that can produce creative work or act creatively. Recent research has looked at how we can make such research more scientifically rigorous. In particular, how do we evaluate how creative these systems are? Various contributions (including my own) have been made in terms of methodologies for evaluation of creative systems; the time is ripe for comparing these methods in different domains. Students may wish to employ comparison metrics such as information-theoretic measures, statistical methods, relating the results to user evaluations and/or computational modelling in this research. It is expected that the PhD work will result in clear recommendations to the computational creativity research community about how to evaluate their systems, and contribute towards solving any issues not addressed by the fledgling existing methodologies that exist.

Supervised by Dr Anna Jordanous*

 

Expressive musical performance software

Traditionally, when computational software performs music the performances can be criticised for being too unnatural, lacking interpretation and, in short, being too mechanical. However much progress has been made within the field of expressive musical performance and musical interpretation expression. Alongside these advances have been interesting findings in musical expectation (i.e. what people expect to hear when listening to a piece of music), as well as work on emotions that are present within music and on how information and meaning are conveyed in music. Each of these advances raises questions of how the relevant aspects could be interpreted by a musical performer. Potential application areas for computer systems that can perform music in an appropriately expressive manner include, for example, improving playback in music notation editors (like Sibelius), or the automated performance of music generated on-the-fly for 'hold' music (played when waiting on hold during phone calls). Practical work exploring this could involve writing software that performs existing pieces, or could be to write software that can improvise, interpreting incoming sound/music and generating an appropriate sonic/musical response to it in real time.

Supervised by Dr Anna Jordanous*

 

Digital preservation of the information within musical/sonic material

Digital preservation of audio material raises many interesting questions to be investigated, including how to archive a sound, what metadata to keep, and future-proofing. Of particular interest is how to explore issues of retention of musical/sonic information from relevant digital audio material, for later access and analysis. Sound and music are typically very open to interpretation, with much information being conveyed through musical/sonic material. Music Information Retrieval (MIR) allows us to see what information is communicated by musical material, using techniques from Computing and Music. Typically MIR is applied to digital rather than physical materials and comes in a variety of forms that could be explored, such as using digital tools or computational analysis for informing and enhancing musicological analysis or musical interpretation. In this PhD project, the PhD candidate will carry out such explorations, towards the development of an archive or a methodology for existing archives to access and retrieve musical information from archive music-based data.

Supervised by Dr Anna Jordanous*

 

Music on the Semantic Web

The Semantic Web is a vision of the Web where items on the web are data, which get linked together if they are data referring to similar things. In the Semantic Web, "a computer program can learn enough about what the data means to process it." (Tim Berners-Lee, Weaving the Web, 2000) There are some data and ontologies (computational models of knowledge) published on the Semantic Web about music, for example the Music Ontology (musicontology.com). Research is starting to emerge on using information retrieval in conjunction with data on the Semantic Web; this project proposes that the PhD candidate explores how Music Information Retrieval (MIR) can be enhanced using Semantic Web data and tools. During this PhD project, the candidate would look at a particular question in music information retrieval, such as how to use MIR to perform computational musicological analysis or how to identify music that is intended to express similar meanings or emotions. (Alternatively the candidate may wish to address a different music information retrieval problem, in an area of specific interest to them; this is welcome.) The PhD candidate would explore how this MIR question can be addressed by using music-specific Semantic Web data/models/technologies to enhance the process of identifying relevant information. It is expected that the PhD candidate will produce computational tools or software that engages directly with the Semantic Web in order to perform the musical information retrieval task. The performance of Semantic-Web enhanced solutions should be compared to traditional MIR solutions for that task, if any exist, and evaluated as to the accuracy and comprehensiveness with which the tools or software carry out the task.

Supervised by Dr Anna Jordanous*

 

Treating Obesity with EEG

Obesity is one of the prime challenges facing in healthcare system with around 20% of the world’s population is affected by this. Obesity is considered as a complex syndrome in which psychological, genetic, environmental and physiological factors involved to develop the obese phenotype. Until now, health policies was ineffective at preventing the increase in obesity rates and this indicates the requirement of developing an effective intervention at individual and population level. There are different arguments regarding the concept of food craving with one view consider food addiction as a social phenotype that is commonly seen in a group of people with obesity which is similar to drug addiction and others think obesity as a result of food addiction with some foods are contain addictive ingredients such as salt, sugar and fat etc. Studies about craving changes in brain were examined by cue response techniques and these activities was identified in caudate, insula and hippocampus. These three parts are reported to be aiding the common substrate hypothesis for drug and food craving. The objective of this research is to use EEG and signal processing to treat obesity.

Supervised by Dr Palaniappan Ramaswamy


Improving attention with sounds

There is much information on the Internet on how sounds can influence and help with concentration. But not much conclusive information on the effects of such sounds has been found, and there is a possibility that the use of these sounds to manipulate the brain is merely an urban myth. As people find themselves easily distracted, these sounds become a form of alternative treatment for them, and it is thus important to investigate these sounds and to what extent they have the intended effects on concentration. This project will study this aim using EEG.

Supervised by Dr Palaniappan Ramaswamy

 

Stress management

The fundamental aspect of human experience is awareness. Combined with the ability to think, imagine and understand it results into the beautiful cosmic play we experience. However, with it comes along a multitude of problems, often illusory in nature – such as stress, anxiety, anger, negativity, etc. It is isn't hard to guess that in such states our behaviour is significantly altered, usually in harmful ways for both – us and the environment. There are techniques such as meditation, music, humour which can help us come back to our “real” senses and feel happier/peaceful again. So the fundamental enquiry would be about what sort of things do help us achieve a happier state, and moreover what's their impact on both short term and long term brain functioning. This project will study this aim using EEG.

Supervised by Dr Palaniappan Ramaswamy

 

Pocket brain-computer interface design for device control

The project will investigate novel algorithms and stimuli design for a practical, real-time and portable brain-computer interface (BCI) design for device control. It will involve electroencephalogram (EEG) data collection using available equipment from volunteers and the analysis will be carried out initially using Matlab software to allow some form of device control. Next, the feasibility of implementing the complete framework in a portable device will be explored. Therefore, familiarity with Matlab and portable device programming (such as Android) will be necessary. Understanding of basic signal processing concepts such as filtering and spectral analysis will be required. In addition, electronic component development skills will be desirable. Knowledge of neural mechanisms and data collection experience will be useful but not mandatory.

Supervised by Dr Palaniappan Ramaswamy

*Please note, Anna is on maternity leave until August 2019, but please contact Colin Johnson if you are interested in one of these projects, who will act on her behalf for PhD applications

School of Computing, University of Kent, Canterbury, Kent, CT2 7NF

Enquiries: +44 (0)1227 824180 or contact us.

Last Updated: 09/04/2019