Comparing information extraction between instance-based data models and relational data models
Bibliographic record
Abstract
Instance-based representation has been developed to overcome the limitations of class- based models for storing data. A class-based data model organizes data into pre- defined classes that represent specific entities within a domain. However, instance- based model introduces two separated layers for representing instance and classes, freeing instances from pre-defined, fixed schemas and enabling more dynamic and exible data representations. Despite the well-established theoretical foundations of instance-based representation, there is little empirical research that investigates its practical usefulness. In this study, we conduct an experiment to compare the effectiveness of information extraction between instance-based data models and class-based data models. Participants randomly received data represented using data structured according to one of the models and answered information extraction/retrieval questions. The results show that, depending on the type of information extraction task, one representation supported more effective retrieval than the other, suggesting that the models can be complementary. In complex use cases including extracting information about relationships of instance/entities and retrieving information involving instances from different classes, the instance-based model outperformed the class- based model. On the other hand, for simpler use cases involving extracting information about cardinalities of relationships and retrieving information involving only one entity (i.e., instances from a same class), the class-based model proved to be more effective. The findings both provide empirical evidence for the effectiveness and usefulness of the instance-based model and demonstrate how it can complement the class-based model in representing the domain.
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 | 0.028 | 0.153 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.005 | 0.006 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.006 | 0.022 |
| Open science | 0.002 | 0.004 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
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".