Detection of Similar Legal Cases on Personal Injury
Bibliographic record
Abstract
The Canadian case system is based on the principle of stare decisis and the concept that like cases should be decided alike. Each judge, when deciding a matter before him or her, selects the prior cases on which to rely. Recently researchers have begun exploring the use of legal text data to find similar cases to assist lawyers with legal research as well as to assist self-represented litigants with legal aid tools. Due to differences in writing style, verbosity, variation in feature importance, case complexity, and subjective bias in judgements, the analysis of legal text using computational models offers interesting challenges for computer scientists. In this study, we explore the problem of finding similar personal injury cases in which plaintiffs claimed compensation specifically for neck and/or back injuries. We extracted and pre-processed unlabeled legal text data and developed deep-learning models across three stages to gradually improve model performance. At each stage, the subset of results was evaluated and validated by a team of lawyers based on qualitative criteria, with the feedback used to refine the model at the next stage. The results demonstrate that semantic similarity between two cases does not ensure that they are legally similar, and artificial intelligence and deep learning techniques for analyzing legal text data can help detect legally similar cases.
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 | 0.001 |
| Bibliometrics | 0.007 | 0.002 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.002 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 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".