The Indigenous Languages Technology (ILT) project at the National Research Council of Canada, and its context
Bibliographic record
Abstract
Recent decades have seen the creation of hundreds of Indigenous language revitalization and reclamation projects across Canada. Most are located inside Indigenous communities; a few are centred in universities. They range in scale from the ambitious, multi-language efforts of the First Peoples’ Cultural Council (FPCC), which is located in and funded by the Province of British Columbia and is active there and elsewhere in Canada (https://fpcc.ca/), to unfunded one- or two-person volunteer efforts in remote communities The project that I have the honour to lead described in this paper – and on this periodically updated website: https://nrc.canada.ca/en/research-development/research-collaboration/programs/canadian-indigenous-languages-technology-project – is managed inside the National Research Council of Canada (NRC), which is the primary research and technology organization of the Government of Canada. However, the reader should not conclude that “settler” (non-Indigenous) governments play a key role in Indigenous language revitalization in Canada. All successful language revitalization projects that I am aware of are Indigenous-run or collaborate closely with Indigenous language activists. Our Indigenous Languages Technology (ILT) project at NRC (henceforth, “NRC-ILT”) is not itself a language revitalization project: it builds tools that people working on language revitalization sometimes find useful. We – the NRC-ILT team – are like lighting technicians or stagehands in a theatre: we are not the actors (we do not appear on stage), nor are we in charge of the production. Those roles are filled by Indigenous language activists and their communities. Language revitalization would go on without us. At times, however, our technical help may make it go slightly better. Everything we have accomplished has been done in collaboration with Indigenous stakeholders.
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 | 0.008 | 0.015 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.004 |
| Science and technology studies | 0.035 | 0.014 |
| Scholarly communication | 0.015 | 0.004 |
| Open science | 0.004 | 0.007 |
| Research integrity | 0.004 | 0.007 |
| Insufficient payload (model declined to judge) | 0.017 | 0.003 |
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".