Colloquium Computer Science - Dr. Rix Groenboom & Leonid Borodaev EMEA-Parasoft
When: | Th 05-03-2020 13:00 - 14:00 |
Where: | 5161.0289 Bernoulliborg |
Title: Machine Learning for software analysis
Abstract:
Static code analysis (SA) refers to a set of compile time techniques aimed at assessing the properties of the run time behaviour of the analysed program. SA is used as a part of the software development process in safety critical applications to establish an absence of errors in the compiled program. The biggest challenge in adopting SA as a part of an organisation’s SDLC is believed to be an abundance of false positives in the report of a typical SA tool and the prohibitively long time that is required to perform a deep analysis of a large project. We propose the use of modern AI approaches to overcome these challenges by speeding up the SA process and reducing the number of false positives in the SA report.