Automated Analysis of Public Health Protection Data
Aberystwyth University, Computer Science
Project ID: AU30035
Annual Stipend: £14,340
Application Deadline: 8 September 2018
Noise pollution is the second highest risk to public health after air pollution. This project will investigate the application of machine learning algorithms to extract additional value from audio and associated data recorded from NoiseApp. The NoiseApp is developed by RHE Global for capturing data (audio recordings, location, time and other meta data) for supporting cases of noise nuisance. It helps both victims and professionals investigating complaints to resolve issues quickly. It has been used to capture a large existing database of related data. This dataset has been largely unexplored in terms of additional value that could be extracted from it. This project will involve the application of supervised and unsupervised methods, signal processing and analysis. Potential applications include automatic identification of the nature / seriousness of the problem (this could include, for example, rapid identification of potential cases of domestic abuse allowing prompt alerting of the relevant authorities), or improvements to the app to include modelling the sound environment by inferring directly from the audio or using vision based methods, or both.
The work involved would include a thorough review of current literature (including available tools and techniques), exploring the available data using existing machine learning tools, signal processing and analysis of sound recordings using traditional and deep learning methods and integration of the two. Some additional manual data annotation (e.g. by human subjects listening to and classifying a subset of the sound files) may be required to maximise the utility of the data.
Depending on the progress made in the initial stages, the second part of the project will investigate estimation of properties of the sound environment. This would likely require the capture of additional sound data under varied and known environments (room shape, clutter, location, type and volume of noise). This would potentially allow more accurate estimation of the seriousness of the problem, where it is coming from etc.
The final part of the work, given sufficient time, could involve tools for estimating properties of the capture location (capture location shape and size) based on visual input (shape from multiple views) or acoustic properties directly (e.g. actively measuring noise impulse response properties).
The prospective applicant should have a minimum of a 1st or good 2:1 in a relevant degree, and be available to take up the studentship by 1st October 2018. The project is part-funded by the European Social Fund (ESF) through the European Union’s Convergence programme administered by the Welsh Government. KESS II PhD scholarships are collaborative awards with external partners. (Applicants need to only apply, they do not need to search for partners.)
To apply, please submit the following to the Postgraduate Admissions Office (address below) by 8 September 2018
- A completed Research Programme Application Form, two references. Application and reference forms may be downloaded from http://www.aber.ac.uk/en/postgrad/howtoapply/
- A completed KESS II Participant proposal form (put the reference number AU30035 in the top right hand box of the application form) and an up-to-date CV. KESS II application forms are available to download at the link below.
- A PhD proposal of up to 1,000 words where you expand on your experience and interests and describe why you are a good candidate for this research studentship. Please refer to the Project Description.
Value of Award: A stipend of £14,340 (rising in accordance with inflation for the remaining two years). Each scholarship has an additional budget for travel, equipment/consumables and training to support your research. KESS II PhD Scholarship holders do not pay fees.
Length: Full-time for 3 years. (Theses must be submitted 6 months after the funded three year study period.)
Training: The achievement of a Postgraduate Skills Development Award (PSDA) is compulsory for each KESS II scholar (The PSDA is based on a 60 credit award, which is an additional award to the PhD).
Eligibility: To be eligible to apply for a KESS II award, you must be resident, upon starting the scholarship, in the Convergence Area of Wales and you must be able to take paid employment in the Convergence area on completion of the scholarship.
The Convergence Area means the following counties of Wales:
Isle of Anglesey
Neath Port Talbot
Rhondda Cynon Taff
For further student eligibility criteria related to the individual projects, please view the details of the individual project above.
Informal enquiries should be made to Bernard Tiddeman at firstname.lastname@example.org or 01970 621777
Address for applications:
Postgraduate Admissions Office
Recruitment & Admissions
Student Welcome Centre
Quote Reference AU30035
Closing date for applications 8 September 2018