Exploring Methods for Linked Data Model Evaluation in Practice
Bibliographic record
Abstract
Ontology development and data modeling are core components of any linked data project. Through our own experiments building a linked data ontology for our collections, we wondered: how are our peers in the linked data community evaluating their ontologies? Are participants engaging in ontology evaluation? What methodologies and evaluation criteria are they using? Are they documenting and sharing their processes? In this paper, we present findings from a survey conducted in the fall of 2018, aimed at professionals from libraries, archives, and museums (LAM) who were part of the data modeling team on linked data projects. The purpose of this survey was to better understand the reality of ontology evaluation in the context of a linked data project. We found that our colleagues were engaging in data modeling as part of linked data projects in a variety of different tasks and roles. There was some ambiguity with respect to evaluation, possibly in part due to the iterative nature of the modeling process. Evaluation is engaged iteratively and informally through use cases, competency questions, and testing of the data in the application. On the whole, not being shared widely outside of a project. The identified barriers to evaluating their models included: lack of knowledge, resources, and documentation.
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 | 0.443 | 0.604 |
| Meta-epidemiology (narrow) | 0.004 | 0.004 |
| Meta-epidemiology (broad) | 0.004 | 0.004 |
| Bibliometrics | 0.019 | 0.014 |
| Science and technology studies | 0.007 | 0.021 |
| Scholarly communication | 0.030 | 0.036 |
| Open science | 0.011 | 0.023 |
| Research integrity | 0.008 | 0.009 |
| Insufficient payload (model declined to judge) | 0.014 | 0.002 |
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".