Large-Scale Empirical Study of Important Features Indicative of Discovered Vulnerabilities to Assess Application Security
Bibliographic record
Abstract
Existing research on vulnerability discovery models shows that the existence of vulnerabilities inside an application may be linked to certain features, e.g., size or complexity, of that application. However, the applicability of such features to demonstrate the relative security between two applications is not well studied, which may depend on multiple factors in a complex way. In this paper, we perform the first large-scale empirical study of the correlation between various features of applications and the abundance of vulnerabilities. Unlike existing work, which typically focuses on one particular application, resulting in limited successes, we focus on the more realistic issue of assessing the relative security level among different applications. To the best of our knowledge, this is the most comprehensive study of 780 real-world applications involving 6498 vulnerabilities. We apply seven feature selection methods to nine feature subsets selected among 34 collected features, which are then fed into six types of machine learning models, producing 523 estimations. The predictive power of important features is evaluated using four different performance measures. This paper reflects that the complexity of applications is not the only factor in vulnerability discovery and the human-related factors contribute to explaining the number of discovered vulnerabilities in an application.
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 | 0.004 | 0.029 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.005 | 0.003 |
| Science and technology studies | 0.000 | 0.001 |
| 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 | 0.000 |
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".