Automatically Generating Release Notes with Content Classification Models
Bibliographic record
Abstract
Release notes are admitted as an essential technical document in software maintenance. They summarize the main changes, e.g. bug fixes and new features, that have happened in the software since the previous release. Manually producing release notes is a time-consuming and challenging task. For that reason, sometimes developers neglect to write release notes. For example, we collect data from GitHub with over 1900 releases, and among them, 37% of the release notes are empty. To mitigate this problem, we propose an automatic release notes generation approach by applying the text summarization techniques, i.e. TextRank. To improve the keyword extraction method of traditional TextRank, we integrate the GloVe word embedding technique with TextRank. After generating release notes automatically, we apply machine learning algorithms to classify the release note contents (or sentences). We classify the contents into six categories, e.g. bug fixes and performance improvements, to represent the release notes better for users. We use the evaluation metric, e.g. ROUGE, to evaluate the automatically generated release notes. We also compare the performance of our technique with two popular extractive algorithms, e.g. Luhn’s and latent semantic analysis (LSA). Our evaluation results show that the improved TextRank method outperforms the two algorithms.
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 | 0.002 | 0.008 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.004 | 0.002 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.002 |
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".