Academic/Research

Department of Computer Science
1 x Research Assistant Ref: 06442/IA
2 x PhD Studentships in the RADAR (Risk Assessment and Decision Analysis Research) Group Ref: 06460/IA

We invite applications for three positions beginning as soon as possible. One is for a Research Assistant (two years fixed term) and the other two are for three-year PhD studentships.

The RA will be engaged in developing new inference techniques and implementing these in Java; however we will consider candidates with strong skills in only one of these areas.

In addition to those major tasks described above the RA's duties will involve: Determining and testing performance and reliability characteristics; Building statistical models to test the new algorithms; Extending existing user interfaces and object oriented architecture and staying abreast of novel academic work in the area.

Applicants for the RA and the PhD studentships positions should:

• Have obtained at least good first degree in a numerate subject (including a good background in computer science, mathematics or probability and statistics);
• Exceptional programming capability;
• Be interested in applying leading-edge research in industry;
• Be capable of self-motivation (although working within a group environment);
• Have good interpersonal and communication skills.

This post is available on a fixed-term contract for 2 years. Salary in the range of �26,736 - �29,768 per annum.

The studentship pays all fees plus a �12,000 per annum maintenance grant. Where appropriate the salary can be supplemented by some consultancy income through Agena and teaching income at Queen Mary.

The RADAR research group in the Computer Science Department at Queen Mary, University of London focuses on improved methods of risk assessment and statistical modelling using Bayesian networks. The group has recently been awarded two new projects. The first is an EPSRC funded project, "DyFusion: Towards a Novel Universal tool for Modelling and Reasoning under Uncertainty", partly sponsored by Agena Ltd. The second project is on "Decision Support for Component-Based Software Testing" and is partly sponsored by Motorola Research Labs.

The DyFusion project aims to produce a general inference algorithm that can handle very general classes of hybrid BN models for high dimensional statistical modeling with complex dependency structures on latent and observable variables, such as multivariate regression or factor analysis and dynamic space state models, such as used to model time series or control systems.

The Motorola sponsored project aims to capture and represent information about the relationship between the reliability of a software system in use and the presence of defects in different 'components' of the software system and to use this relationship to make predictions relevant to the software testing process. The primary means of modelling and inference will be based on established Bayesian network technology.

Details about the research projects can be found here:
www.dcs.qmul.ac.uk/~martin/

Details about the RADAR group can be found here:
www.dcs.qmul.ac.uk/~norman/radarweb

Informal enquires can be made by contacting Dr Martin Neil at martin@dcs.qmul.ac.uk.

Please quote the correct reference number when applying for these posts and send them to: suew@dcs.qmul.ac.uk

or by post to: Mrs Sue White, Department of Computer Science, Queen Mary, University of London, London, E1 4NS, UK.

For more information please visit http://www.dcs.qmul.ac.uk/staff/recruit/

A downloadable application form can be found at http://www.hr.qmul.ac.uk/vacancies/

Applications should include a covering letter detailing a description of relevant practical experience, a full CV, an application form and the names and email addresses of two referees.

The closing date for applications is Friday January 26th 2007.

Working towards equal opportunities