The Gap between Deep Learning and Law: Predicting Employment Notice.
Bibliographic record
Abstract
This study aims to determine whether Natural Language Processing with deep learning models can shed new light on the Canadian calculation system for employment notice. In particular, we investigate whether deep learning can enhance the predictability of notice period, that is, whether it is possible to predict notice period with high accuracy. A major challenge with the classification of reasonable notice is the inconsistency of the case law. As argued by the Ontario Court of Appeal, the process of determining reasonable notice is "more art than science". In a previous study, we assessed the predictability of reasonable notice periods by applying statistical machine learning to a hand-annotated dataset of 850 cases. Building on this past study, this paper utilizes state-of-the-art deep learning models on a free-text summary of cases. We further experiment with a variety of domain adaptations of state-of-the-art pretrained BERT-esque models. Our results appear to show that the domain adaptations of BERT-esque models negatively affected performance. Our best performing model was an out-of-the-box RoBERTa base model which achieved a 69% accuracy using a +/-2 prediction window.
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 | 0.013 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.002 | 0.002 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.002 | 0.003 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.004 | 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".