Automatic Domain-Specific Corpora Generation from Wikipedia - A Replication Study
Bibliographic record
Abstract
Replication studies help mature our knowledge and attempt to validate the findings of a prior piece of research. However, these studies are still rare in the Requirements Engineering field. Additionally, the rapidly advancing realm of Natural Language Processing (NLP) is creating new opportunities for efficient, machine-assisted workflows application which can bring new perspectives and results to the forefront. Thus, in this paper, we replicate and extend a previous study (baseline), a tool, WikiDoMiner, which automatically generated domain-specific corpora by crawling Wikipedia. In this study, we investigated and executed the implementation of WikiDoMiner (open-sourced code from the original paper) to recreate the results. This allowed us to strengthen the external validity of the original study. We extended the baseline to evaluate additional data sets and generated nuanced results using state-of-the-art NLP techniques such as Bidirectional Encoder Representations from Transformers (BERT). Results showed that due to the growing content in Wikipedia, the corpus generated for the Railways and Networks domains did not precisely match the results from the baseline. However, utilizing the state-of-the-art KeyBERT library from the Huggingface AI community enhanced the results, eventually generating a meaningful corpus compared to the baseline.
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 | 0.024 | 0.096 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.002 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.002 | 0.004 |
| Open science | 0.002 | 0.003 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.003 | 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".