Evaluating the Effectiveness of Code2Vec for Bug Prediction When Considering That Not All Bugs Are the Same
Bibliographic record
Abstract
Bug prediction is an area of research focused on predicting where in a software project \nfuture bugs will occur. The purpose of bug prediction models is to help companies spend \ntheir quality assurance resources more efficiently by prioritizing the testing of the most \ndefect prone entities. Most bug prediction models are only concerned with predicting \nwhether an entity has a bug, or how many bugs an entity will have, which implies that all \nbugs have the same importance. In reality, bugs can have vastly different origins, impacts, \npriorities, and costs; therefore, bug prediction models could potentially be improved if they \nwere able to give an indication of which bugs to prioritize based on an organization’s needs. \nThis paper evaluates a possible method for predicting bug attributes related to cost by \nanalyzing over 33,000 bugs from 11 different projects. If bug attributes related to cost can \nbe predicted, then bug prediction models can use the approach to improve the granularity of \ntheir results. The cost metrics in this study are bug priority, the experience of the developer \nwho fixed the bug, and the size of the bug fix. First, it is shown that bugs differ along each \ncost metric, and prioritizing buggy entities along each of these metrics will produce very \ndifferent results. We then evaluate two methods of predicting cost metrics: traditional deep \nlearning models, and semantic learning models. The results of the analysis found evidence \nthat traditional independent variables show potential as predictors of cost metrics. The \nsemantic learning model was not as successful, but may show more effectiveness in future \niterations.
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 | 0.003 | 0.012 |
| Meta-epidemiology (narrow) | 0.003 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.001 |
| Bibliometrics | 0.004 | 0.003 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.004 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.002 | 0.003 |
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".