The Department of Mathematics and Information Sciences at Northumbria University have a funded PhD studentship available to start October 2014.

Project title: Developing novel statistical methodology for interpreting forensic textile fibre evidence

Supervisors: Dr Guangquan Li, Dr Kelly Sheridan and Dr Ray Palmer

Project Description: Textile fibres often become crucial as forensic evidence in criminal investigations to demonstrate links between crime scenes, suspects and victims, providing intelligence information to the Police. To help decide whether or not a questioned fibre could have come from a putative source, various measuring techniques are applied, from measuring the shape and dimensions of the fibres to utilising UV-visible microspectrophotometry for colour absorbance spectra analysis. However, interpretation of such information relies on experts’ experience and is highly subjective. Although it is unusual for forensic fibre experts to disagree on such matters, it does happen as in the recent murder case appeal (R-v-Simon John Hall), where (dis)agreement on fibre evidence became critical.

In collaboration with Dr Sheridan and Dr Palmer from the Centre for Forensic Science at Northumbria, the goal of this interdisciplinary research project is to develop an objective and unified statistical approach for interpreting fibre evidence. To achieve this goal, the project will first investigate the suitability of utilising multivariate techniques such as principal component analysis and linear discriminant analysis for determining the ‘matching probabilities’ for fibres. It is valuable to reveal the strengths and limitations of these techniques as they have not been fully explored in casework scenarios. The main anticipated challenge for these methods would be to incorporate various sources of information coherently. A novel statistical method will hence be developed to encapsulate various fibre measurements. This model will be placed within the Bayesian framework, which offers the flexibility in accounting for multiple measurements with a key advantage of formally quantifying uncertainty in results.

Informal Enquiries: Enquiries regarding this studentship should be made to: [log in to unmask]

Eligibility criteria: Applicants should hold a first or upper second class honours degree (in a relevant subject) from a British higher education institution, or equivalent. Students who are not UK/EU residents are eligible to apply, provided they hold the relevant academic qualifications, together with an IELTS score of at least 6.5.

To apply, e-mail [log in to unmask] to request the application form, quoting the advert reference EE/DRFCEIR7/53588. Alternatively, use the application link https://northumbria.hobsons.co.uk/emtinterestpage.aspx?ip=grad.

Deadline for applications: 07 April 2014

Start Date: 1 October 2014

Funding Notes: The studentship includes a full stipend, paid for three years at RCUK rates (in 2013/14 this is 13,863 pa) and Home/EU fees. Overseas candidates are also eligible to apply.

References:

K. Wiggins, R. Palmer et al., An investigation into the use of calculating the first derivative of absorbance spectra as a tool for forensic fibre analysis, Sci. Justice 47 (1) (2007) 9-18

C. Adam, et al., Classification and Individualisation of Black Ballpoint Pen Inks Using Principal Component Analysis of UV--vis Absorption Spectra, Forensic Science International 174(1) (2008) 16-25

G. Li, et al., BaySTDetect: detecting unusual temporal patterns in small area data via Bayesian model choice. Biostatistics, 13 (4) (2012) 695-710.

You may leave the list at any time by sending the command

SIGNOFF allstat

to [log in to unmask], leaving the subject line blank.