10 Tips on Mining Fact Witness Deposition Transcripts in Forensic Investigations
Bibliographic record
Abstract
Fact witness deposition testimony is one of many informational resources available to forensic liability experts in revealing the facts that will guide their investigation and be relied upon in expressing public statements, expert opinions, and expert testimony. However, mining fact witness deposition transcripts for relevant facts can be challenging, and may lead some experts to either dismiss, ignore, or overlook important facts. The consequence of working with incomplete facts may adversely impact an investigation and undermine an expert’s credibility. This paper surveyed approximately 20,000 pages of fact witness deposition transcripts from more than 200 depositions in more than 75 lawsuits in liability and personal injury matters. The results of this survey demystify the transcripts by offering familiarity with the deposition process and providing 10 tips on mining fact witness deposition transcripts. The 10 tips are: (1) request for the transcripts, exhibits, and errata; (2) navigating the transcripts; (3) breaking down the incident-related facts; (4) attentiveness to key words; (5) attentiveness to volunteered information; (6) attentiveness to false and misleading information; (7) attentiveness to corrections in testimony; (8) attentiveness to other relevant documents; (9) attentiveness to questions that can educate the expert; and (10) attentiveness to authenticated photographic exhibits. These tips benefit both seasoned experts, who may only have a limited working knowledge of deposition transcripts, and novice experts, while also providing retaining attorneys with insights into the forensic investigation process.
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 | 0.023 | 0.145 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.007 | 0.005 |
| Science and technology studies | 0.003 | 0.001 |
| Scholarly communication | 0.004 | 0.008 |
| Open science | 0.002 | 0.007 |
| Research integrity | 0.002 | 0.003 |
| Insufficient payload (model declined to judge) | 0.019 | 0.012 |
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".