The 6th International Conference on Predictive Models in Software Engineering
Co-located with ICSM'10, at Timisoara, Romania
Sept 12-13, 2010
Home | Program | Venue | Registration | Call for papers | Dates | Committees | Keynotes | Submit | Promote
How SBSE can Support Construction and Analysis of Predictive Models
- Search Based Software Engineering (SBSE) is an approach to software engineering in which search based optimization algorithms are used to identify optimal or near optimal solutions and to yield insight.
- SBSE techniques can cater for multiple, possibly completing objectives and/ or constraints and applications where the potential solution space is large and complex.
- This talk will provide a brief overview of SBSE, explaining some of the ways in which it has already been applied to construction of predictive models. There is a mutually beneficial relationship between predictive models and SBSE. The talk will outline some novel ways in which SBSE might be used for predictive model construction and how predictive models also have role to play in improving SBSE.
About the speaker:
- Mark Harman is professor of Software Engineering in the Department of Computer Science at King's College London. He is widely known for work on source code analysis and testing and he was instrumental in the founding of the field of Search Based Software Engineering, a field that currently has active researchers in 24 countries and for which he has given 14 keynote invited talks.
- Professor Harman is the author of over 150 refereed publications, on the editorial board of 7 international journals and has served on 90 programme committees.
- He is director of the CREST centre at King's College London. More details are available from the CREST website (http://crest.dcs.kcl.ac.uk/).
Overcoming scale and privacy issues in industrial software studies and repositories
- Replicating a software engineering study to predict effort and defects relies on having public data, preferably in a public repository. Large companies like Microsoft certainly conduct such studies to drive their business, but they cannot easily release their data about bug fixes and customer usage. However, increasingly, the methods and tools that are used to analyze the data are coming from the research community, and therefore can be opened up to wider scrutiny.
- This talk will look at two new tools out of Microsoft, and aspects of the repositories they use. The first is Crane, which uses complex modeling to predict which changes in large code bases will be more likely to produce faults. The second is Web N-Gram which provides cloud based services to drive discovery and innovation based on all the data currently indexed by Bing in the US. Issues of scale and privacy are discussed, as well as how these tools can be integrated into existing research efforts.
About the speaker:
- Judith Bishop is Director of Computer Science in External Research at Microsoft Research, based in Redmond, USA. Her role is to create strong links between Microsoft's research groups and universities globally, through encouraging projects, supporting conferences and engaging directly in research.
- After doing her undergraduate degrees at Rhodes and Natal in South Africa, Dr. Bishop received her PhD from the University of Southampton in 1977 on the relationship of languages to computer architecture. She has over 90 publications including 15 books on programming languages that are available in six languages and read worldwide. Judith serves frequently on international editorial, programme and award committees, and has received numerous awards and distinctions, in particular the IFIP Outstanding Service Award in 2009.