An Intelligent Class – The Sequel: The Development Of A Novel Context Capturing Method For The Functional Auto Classification Of Records
Bibliographic record
Abstract
The need to accurately classify records is a core problem in many domains. Historically, the classification of records was done manually as records were received and then categorized. Unfortunately, due to a significant growth in the volume of records, the need for robust auto-classification methods that can effectively “read” and classify records, is high. Today, significant challenges remain with the development of effective auto-classification processes for records. This is because the records traditionally require functional classification based on context, not topic classification based on content. Functional classification traditionally has been a challenge for both humans and machines, with little research on how to classify a record effectively functionally. To move research forward, this paper will address the challenges of both human and machine classification of records.Firstly, this paper will seek to evaluate the efficacy of human manual classifiers on a classification task, using knowledge from this process to articulate a process for automated functional classification that utilizes a record’s archival diplomatic context. Secondly, this paper will compare the efficacy of manual versus machine (i.e., auto-classification) using a record set with over 500,000 records, using a novel auto-classification approach that leverages a record’s context, not just its content, to improve classification accuracy. As this paper will discuss, there is significant variance between expert human (i.e., records managers) during the manual classification process, with statistically significant differences in their ability to accurately classify both administrative and operational records. Moreover, this paper will demonstrate that an auto-classifier, when trained using key elements of context, can statistically outperform a group of expert human classifiers on a classification task.
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 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 | 0.002 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| 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 itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, 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".