Bibliographic record
Abstract
Knowledge-rich applications can see significant performance improvementsby using domain-specific Knowledge bases (KBs). Populating and enriching these KBs has, thus, become an important challenge. In this thesis, we examine a powerful approach for KB population that is based on knowledge exchange, the process of translating knowledge from one KB to another, even when these KBs use very different concepts, properties, and graph structure to represent their knowledge. We introduce Kensho. A tool for generating mapping rules between two Knowledge Bases. In the data exchange problem, data that is structured under a source schema is transformed into an instance of a target schema. This is accomplished using a set of rules (called mapping rules) that specify the relationship between the source and target schemas. Kensho can produce mapping rules even in the presence of cycles, incompleteness in the source, and in settings with missing or unknown correspondences between properties or property paths. In addition, Kensho performs knowledge translation using value invention to preserve the proper grouping of data in the target KB. We also introduce two tools (Vizcurator and Sassho) that we have created to help domain experts in the task of knowledge translation. Vizcurator aims to help a domain expert understand and curate the source of exchange. Sassho aims to help a domain expert compare and understand mapping rules which are automatically created using a mapping generation tool such as Kensho. Sassho enables a domain expert to create examples that can be used to understand subtle differences among alternative mapping rules and explore the affect of those differences on the data being exchanged. As interest in supporting data exchange between heterogeneous knowledge bases (KBs) has increased, so has interest in benchmarking KB exchange systems. We introduce a set of new requirements for a KB exchange benchmark based on unique characteristics of KBs and based on important lessons learned from other data exchange systems. The field of exchanging information among KBs is relatively new. We outline an extensive research agenda for Knowledge Exchange based our experience in bringing data exchange to knowledge graphs.
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 | 0.003 | 0.018 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.005 | 0.005 |
| Science and technology studies | 0.002 | 0.002 |
| Scholarly communication | 0.008 | 0.007 |
| Open science | 0.003 | 0.009 |
| Research integrity | 0.003 | 0.003 |
| Insufficient payload (model declined to judge) | 0.190 | 0.137 |
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".