An Exploratory Study on Automatic Architectural Change Analysis Using Natural Language Processing Techniques
Bibliographic record
Abstract
Continuous architecture is vital for developing large, complex software systems and supporting continuous delivery, integration, and testing practices. Researchers and practitioners investigate models and rules for managing change to support architecture continuity. They employ manual techniques to analyze software change, categorizing the changes as perfective, corrective, adaptive, and preventive. However, a manual approach is impractical for analyzing systems involving thousands of artefacts as it is time-consuming, labor-intensive, and error-prone. In this paper, we investigate whether an automatic technique incorporating free-form natural language text (e.g., developers' communication and commit messages) is an effective solution for architectural change analysis. Our experiments with multiple projects showed encouraging results for detecting architectural messages using our proposed language model. Although architectural change categorization for the preventive class is moderate, the outcome for the random dataset is insignificant in general (around a 45% F1 score). We investigated the causes of the unpromising outcome. Overall, our study reveals that our automated architectural change analysis tool would be fruitful only if the developers provide considerable technical details in the commit messages or other text.
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 | 0.015 | 0.070 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.003 | 0.002 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.004 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| 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".