The IT Certification Resource Center

Featured Deal

Get CompTIA, Cisco, and Microsoft training courses free for a week.
Learn More ❯

BCS announces distinguished dissertation winner

5 November 2014

Juan Reutter, who studied for his PhD in the School of Informatics at the University of Edinburgh, has been selected by the Council of Professors and Heads of Computing (CPHC) in conjunction with BCS, the Chartered Institute for IT, as the winner of the distinguished dissertation competition 2014.

 

Juan’s dissertation entitled: Graph Patterns: Structure, Query Answering and Applications in Schema Mappings and Formal Language Theory, was selected as the overall winning dissertation from more than 25 entries from universities across the UK. Simon Poulding from the University of York was selected as runner up for his dissertation entitled: The Use of Automated Search in Deriving Software Testing Strategies.

 

Simon Dobson, Professor of Computer Science at the University of St Andrews and Chair of the 2014 Distinguished Dissertations panel says: “I’d like to congratulate the authors of these theses, their supervisors and departments. This year’s entries were of particularly high quality and it was a pleasure to see the diversity of research being undertaken by post-graduate students in the UK today.”

 

The annual award selects the best British PhD/DPhil dissertations in computer science with the aim of making more visible the significant contribution made by the UK - in particular by post-graduate students - to computer science. The winning dissertation and runner-up submissions will be published on the BCS website. The prize winner will also have their dissertation published in hard copy which will be available later this month via the BCS bookshop. They will receive their award at the 2014 BCS Roger Needham Lecture (19 November) at the Royal Society, London.

 

Sally Smith, Chair of CPHC and Head of the School of Computing at Edinburgh Napier University says: “CPHC continues to support the Distinguished Dissertation competition because we believe that the work being undertaken by post graduates in UK universities is very important to the continued development of research in computer science. My congratulations to all who were submitted for consideration and I wish all the best for your future careers."

 

Juan Reutter says of his dissertation: “There is an initiative right now called the semantic web, which aims to transform the internet into a "web of data" that would allow data to be shared in the same way we are able to share webpages and links now. The problem is that this web of data contains a huge amount of databases of very different nature, all of them linked together. My research provides a framework to understand how to extract information from these databases once they are integrated, or summarised, into a single database.”

 

Simon Poulding says of his dissertation: “Many of the products we use on a daily basis have highly complex software at their heart, and companies must use sophisticated testing methods if they are to avoid diverting all their efforts to ensuring the reliability and safety of this software. This thesis makes a small contribution by demonstrating how sophisticated testing strategies may be automated so that high quality software can be produced using fewer resources. The less money that is spent on finding defects in software, the more can be used to create the next generation of cutting-edge products.”

The dissertations are available online

 

Further details about the dissertations:

Winner - Juan Reutter, University of Edinburgh 
Short summary: Graph data management is currently one of the most active topics in knowledge engineering, driven by the graph models that underpin the Semantic Web, social network analysis, biological databases, geographical information systems, and many others.

 

One of the key problems in the area is to understand how to efficiently query huge volumes of graph-based data. Essentially the problem is to find a sub-graph pattern representing the information being sought. This is a challenge in itself, made more complicated by the fact that real-world queries often lead to incomplete graph patterns. The work presented in this thesis shows that graph pattern queries are computationally more challenging than other common paradigms for knowledge retrieval. It develops restrictions on patterns to form a tractable sub-set that can be used for efficient queries, and applies these results to problems in information exchange.

 

The committee felt that this thesis was exceptionally strong, containing both theoretical analysis and practical applications to an area of great and growing importance in computer science. It provides a well-founded way of addressing a topic that spans a large range of fields of current interest in a way that offers significant prospects for further fruitful research.

 

About the author: Juan Reutter is an assistant professor at the Department of Computer Science at the Pontificia Universidad Católica de Chile, and an associate investigator of the Centre for Semantic Web Research. He studied for his PhD in the School of Informatics at the University of Edinburgh, working in the database group of the Laboratory for Foundations of Computer Science under the supervision of Professor Leonid Libkin.

 

His research interests are in different aspects of data management and automata theory, including graph and semi-structured databases, ontology-based data access, data exchange, metadata management and the semantic web. He received the Best Paper Award at PODS in 2011 and the “Ramon Salas” Award for the best work in engineering produced by Chilean researchers in 2012. He has served on the program committee for SIGMOD and AAAI conferences and has published several papers in major database conferences and journals.

 

Runner-up - Simon Poulding, University of York 
Short summary: Some strategies for selecting test data are, in theory, very effective at detecting faults in software but, in practice, are difficult and costly to use. This thesis proposes that metaheuristic search may be used to bridge this gap between theory and practice, and demonstrates how such an approach enables the use of a sophisticated and effective strategy for random test data generation that would otherwise be too costly. The cost-effectiveness of this search-based approach is enhanced by the rigorous empirical determination of an efficient configuration of the search algorithm, and by the exploitation of commodity GPU cards to parallelise the algorithm. The use of a flexible grammar representation for the domain of test inputs ensures that this approach can be applied to a wide range of software.

 

The committee felt that the thesis takes an approach that has been regarded as attractive in theory but difficult in practice, and turns it into a potentially useful tool for testing large systems. This has enormous implications for the practice of software engineering and for the development of robust and dependable software under realistic industrial conditions.

 

About the author: Simon Poulding studied for his PhD in the Department of Computer Science at the University of York.  During this time he participated in two inter-university research projects on search-based software engineering (SEBASE and DAASE) that were funded by the EPSRC. Simon previously worked as a commercial software engineer before returning to academia to attain an MSc in Mathematics prior to starting his PhD. He is currently a postdoctoral researcher in the Software Engineering Research Lab at the Blekinge Institute of Technology, Sweden, where he is investigating techniques to verify the performance efficiency and robustness of software systems.