Tweaking Association Rules to Optimize Software Change Recommendations
Bibliographic record
Abstract
Past researchs have been trying to recommend artifacts that are likely to change together in a task to assist developers in making changes to a software system, often using techniques like association rules. Association rules learning is a data mining technique that has been frequently used to discover evolutionary couplings. These couplings constitute a fundamental piece of modern change prediction techniques. However, using association rules to detect evolutionary coupling requires a number of configuration parameters, such as measures of interest (e.g. support and confidence), their cut-off values, and the portion of the commit history from which co-change relationships will be extracted. To accomplish this set up, researchers have to carry out empirical studies for each project, testing a few variations of the parameters before choosing a configuration. This makes it difficult to use association rules in practice, since developers would need to perform experiments before applying the technique and would end up choosing non-optimal solutions that lead to wrong predictions. In this paper, we propose a fitness function for a Genetic Algorithm that optimizes the co-change recommendations and evaluate it on five open source projects (CPython, Django, Laravel, Shiny and Gson). The results indicate that our genetic algorithm is able to find optimized cut-off values for support and confidence, as well as to determine which length of commit history yields the best recommendations. We also find that, for projects with less commit history (5k commits), our approach produced better results than the regression function proposed in the literature. This result is particularly encouraging, because repositories such as GitHub host many young projects. Our results can be used by researchers when conducting co-change prediction studies and by tool developers to produce automated support to be used by practitioners.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.005 | 0.028 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.001 |
| Bibliometrics | 0.004 | 0.002 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.003 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".