Mining Software Repositories with a Collaborative Heuristic Repository
This addresses the issue of categorization errors impacting software engineering studies, offering a collaborative solution for researchers, though it appears incremental as it builds on existing weak supervision techniques.
The paper tackles the problem of imprecise categorization of software engineering artifacts by proposing a collaborative heuristic repository that leverages weak supervision to train high-quality classifiers, resulting in improved precision for tasks like commit classification.
Many software engineering studies or tasks rely on categorizing software engineering artifacts. In practice, this is done either by defining simple but often imprecise heuristics, or by manual labelling of the artifacts. Unfortunately, errors in these categorizations impact the tasks that rely on them. To improve the precision of these categorizations, we propose to gather heuristics in a collaborative heuristic repository, to which researchers can contribute a large amount of diverse heuristics for a variety of tasks on a variety of SE artifacts. These heuristics are then leveraged by state-of-the-art weak supervision techniques to train high-quality classifiers, thus improving the categorizations. We present an initial version of the heuristic repository, which we applied to the concrete task of commit classification.