Petra Selmer

PhD student (part-time), Oct 2008 - present
School of Computer Science and Information Systems
Birkbeck College, Malet Street, London, WC1E 7HX

Supervisors: Professor Alexandra Poulovassilis and Dr. Peter Wood
Address:
London Knowledge Lab
23-29 Emerald Street
London WC1N 3QS
Email: lselm01@dcs.bbk.ac.uk
Phone: 020 7763 2101

Biography

Education:

BSc (Computer Science and Informatics): Rand Afrikaans University, 1996 - Johannesburg, South Africa
MSc (Advanced Information Systems): Birkbeck, University of London, 2007

Professional:

Since 1997, I have worked with a large range of companies from global corporations through to small firms, developing solutions including (among many others) Content Management Systems; training systems; e-commerce sites; a comprehensive supply chain management system; systems for managing clinical audit and trials; and a configurable knowledge-base system used for data cleansing and validation.

I am currently working at the Wellcome Trust Centre for Neuroimaging at UCL, where I am developing a system allowing researchers to manage large amounts of complex heterogeneous data.


Research

My PhD is in the area of flexible querying of graph-structured data. In particular, my research involves the novel combination of two hitherto disparate flexible querying methods - query approximation and query relaxation - which return results to the user in a ranked order. In addition to this, the optimisation of the evaluation of such queries shall also be investigated.

Flexible query answering and ranking – due to query relaxation and approximation techniques - is useful in many scientific (such as bioinformatics and astronomy), medical, financial and educational areas where there are large volumes of complex, heterogeneous data. Moreover, the development of techniques to integrate different heterogeneous data sources in these areas has given rise to the situation whereby the users of such data may not be wholly familiar with the underlying structure of the data - even though this ‘pooled’ data, in all likelihood, contains much valuable and useful information for them. Therefore, my research topic shall potentially benefit researchers working with these kinds of data. Additionally, my research topic provides a good opportunity to leverage the benefits of using ontology information with regard to applying semantic knowledge to aid users in querying complex, heterogeneous data.


Publications


Talks and Presentations


Interests

Graph data; NOSQL technologies, particularly graph databases; Semantic Web (Web 3.0) technologies; heterogeneous data integration; query and data processing optimisation; and knowledge-based systems.