Automated Self-Admitted Technical Debt Tracking at Commit-Level: A Language-independent Approach
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Software and systems traceability is essential for downstream tasks such as data-driven software analysis and intelligent tool development. However, despite the increasing attention to mining and understanding technical debt in software systems, specific tools for supporting the track of technical debts are rarely available. In this work, we propose the first programming language-independent tracking tool for self-admitted technical debt (SATD) – a sub-optimal solution that is explicitly annotated by developers in software systems. Our approach takes a git repository as input and returns a list of SATDs with their evolution actions (created, deleted, updated) at the commit-level. Our approach also returns a line number indicating the latest starting position of the corresponding SATD in the system. Our SATD tracking approach first identifies an initial set of raw SATDs (which only have created and deleted actions) by detecting and tracking SATDs in commits’ hunks, leveraging a state-of-the-art language-independent SATD detection approach. Then it calculates a context-based matching score between pairs of deleted and created raw SATDs in the same commits to identify SATD update actions. The results of our preliminary study on Apache Tomcat and Apache Ant show that our tracking tool can achieve a F1 score of 92.8% and 96.7% respectively.
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.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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 it