Natural language processing and machine learning as practical toolsets for archival processing
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Purpose This study aims to provide an overview of recent efforts relating to natural language processing (NLP) and machine learning applied to archival processing, particularly appraisal and sensitivity reviews, and propose functional requirements and workflow considerations for transitioning from experimental to operational use of these tools. Design/methodology/approach The paper has four main sections. 1) A short overview of the NLP and machine learning concepts referenced in the paper. 2) A review of the literature reporting on NLP and machine learning applied to archival processes. 3) An overview and commentary on key existing and developing tools that use NLP or machine learning techniques for archives. 4) This review and analysis will inform a discussion of functional requirements and workflow considerations for NLP and machine learning tools for archival processing. Findings Applications for processing e-mail have received the most attention so far, although most initiatives have been experimental or project based. It now seems feasible to branch out to develop more generalized tools for born-digital, unstructured records. Effective NLP and machine learning tools for archival processing should be usable, interoperable, flexible, iterative and configurable. Originality/value Most implementations of NLP for archives have been experimental or project based. The main exception that has moved into production is ePADD, which includes robust NLP features through its named entity recognition module. This paper takes a broader view, assessing the prospects and possible directions for integrating NLP tools and techniques into archival workflows.
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.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.004 | 0.003 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.002 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
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 it