Towards Utilizing Natural Language Processing Techniques to Assist in Software Engineering Tasks
Bibliographic record
Abstract
Machine learning-based approaches have been widely used to address natural language processing (NLP) problems. Considering the similarity between natural language text and source code, researchers have been working on applying techniques from NLP to deal with code. On the other hand, source code and natural language are by nature different. For example, code is highly structured and executable. Thus, directly applying the NLP techniques may not be optimal, and how to effectively optimize these NLP techniques to adapt to software engineering (SE) tasks remains a challenge. Therefore, to tackle the challenge, in this dissertation, we focus on two research directions: 1) distributed code representations, and 2) logging statements, which are two important intersections between the natural language and source code. For distributed code representations, we first discuss the limitations of existing code embedding techniques, and then, we propose a novel approach to learn more generalizable code embeddings in a task-agnostic manner. For logging statements, we first propose an automated deep learning-based approach to automatically generate accurate logging texts by translating the related source code into short textual descriptions. Then, we make the first attempt to comprehensively study the temporal relations between logging and its corresponding source code, which is later used to detect issues in logging statements. We anticipate that our study can provide useful suggestions and support to developers in utilizing NLP techniques to assist in SE tasks.
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 | 0.005 | 0.029 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.004 | 0.003 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.003 | 0.010 |
| Open science | 0.002 | 0.003 |
| Research integrity | 0.002 | 0.004 |
| Insufficient payload (model declined to judge) | 0.003 | 0.003 |
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".