SOURCE-LINKED INTELLIGENCE

Practical, Learning-Based Tools for Finding and Fixing Bugs

CORDIS · observation · Publication date unknown

ustry. The LearnBugs ERC project, on which this proposal is based, has developed ground-breaking techniques to automatically find bugs and to propose suitable bug fixes. These techniques are based on artificial intelligence and deep learning, making them particularly powerful for kinds of bugs missed by traditional software developer tools. However, these techniques are currently only available as research prototypes, and there is a gap to be bridged in order to integrate them successfully into the software development workflow. This Proof of Concept proposal, named BugGPT, aims to make learning-based techniques for finding and fixing software bugs practical and usable by software developers. The project will develop practical tools that enable software developers to automatically find and fix bugs in their code. To this end, we will perform technical development activities that address the questions of where, when, and how to suggest bug fixes. Furthermore, we will perform business development activities to identify potential customers, to evaluate the usefulness of our tools, and t

Read original source ↗ Open in workspace

recordType

award

status

SIGNED

region

EU

value

150000

unit

EUR