Dataset Characteristics for Reliable Code Authorship Attribution
Bibliographic record
Abstract
Code authorship attribution aims to identify the author of software source code according to the author’s unique coding style characteristics. The lack of benchmark data in the field, forced researchers to employ various resources that often did not reflect real programming practices. Throughout the years, research studies have used textbook examples, students’ programming assignments, faculty code samples, code from programming competitions and files retrieved from open-source repositories as research objects. The diversity of the data raised concerns about the feasibility of capturing the appropriate data characteristics to reliably evaluate code attribution. In this paper, we investigate these concerns and analyze the effect of the dataset characteristics and feature elimination techniques on the accuracy of code attribution. Unlike the majority of the work done in this field, which mainly concentrates on designing new features, we explore the nature of the data used in previous studies and assess the factors that influence the attribution task. Within this analysis, we investigate the robustness of three feature sets regarded as reliable benchmarks in the attribution research. Based on our findings, we define a process for deriving a reduced set of features for accurate and predictable attribution and make recommendations on the dataset characteristics.
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 | 0.022 | 0.161 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.005 | 0.007 |
| Science and technology studies | 0.002 | 0.001 |
| Scholarly communication | 0.004 | 0.004 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.002 | 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".