Leveraging an Enhanced CodeBERT-Based Model for Multiclass Software Defect Prediction via Defect Classification
Bibliographic record
Abstract
Ensuring software reliability through early-stage defect prevention and prediction is crucial, particularly as software systems become increasingly complex. Automated testing has emerged as the most practical approach to achieving bug-free and efficient code. In this context, machine learning-driven methods, especially those leveraging natural language models, have gained significant traction for developing effective techniques. This paper introduces a novel framework for automating software defect prediction, focusing on eight specific defects: SIGFPE, NZEC, LOGICAL, SYNTAX, SIGSEGV, SIGABRT, SEMANTIC, and LINKER. Our research involves a specialized dataset comprising nine classes, including eight common programming errors and one error-free class. The goal is to enhance software testing and development processes by identifying defects within code snippets. The proposed framework utilizes a CodeBERT-based algorithm for defect prediction, optimizing model hyperparameters to achieve superior accuracy. Comparative analysis against established models such as RoBERTa, Microsoft CodeBERT, and GPT-2 demonstrates that our approach yields significant improvements in prediction performance, with accuracy gains of up to 20% and 7% respectively in binary and multi class experimentation. Empirical studies validate the effectiveness of neural language models like CodeBERT for software defect prediction, highlighting substantial advancements in software testing and development techniques. These findings underscore the potential benefits of incorporating advanced machine learning models into the software development lifecycle.
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 | 0.005 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.002 | 0.002 |
| 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".