Performance Evaluation of Transformer-based NLP Models on Fake News Detection Datasets
Bibliographic record
Abstract
Fake news has become a major concern due to its spread on social media. To combat this, various machine learning (ML) techniques have been proposed. However, there is a lack of research on the performance of transformer models using datasets from a wide range of domains. This paper investigates the performance of ML algorithms on three fake news datasets: LIAR, FNC-1 and Balanced Dataset for Fake News Analysis. Pretrained transformer language models such as BERT, RoBERTa, ALBERT and DistilBERT were chosen for this paper. The performance of the models was consistent across all datasets. RoBERTa obtained an accuracy of 69% when trained on the LIAR dataset, an 11% improvement over the existing traditional and deep learning ML model implementations, and an accuracy of 97% when trained on the FNC-1 dataset, proving to be the best-performing model across all the fake news detection datasets utilized in the experiments. DistilBERT trains at a significantly faster rate than the other three variants. The experimental results from the paper can help the research community to continue investigating and gain insights into fake news detection.
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 | 0.011 |
| Meta-epidemiology (narrow) | 0.002 | 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 | 0.003 |
| Open science | 0.002 | 0.001 |
| 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".