An Event-Driven Approach for Querying Graph-Structured Data Using Natural Language.
Bibliographic record
Abstract
An ideal way for people to query graph-based knowledge, including triplestores in the semantic web, would be for them to ask questions in a natural language (NL). However, existing NL query interfaces to graph-based data have limited expressive power and cannot accommodate arbitrarilynested quantification (i.e. phrases such as “a gangster who joined every gang”) together with multiple complex prepositional phrases, such as “in a city located in Illinois in 1918 using a set of keys that was stolen from a gangster”. It would appear that the commonly-used “entity-based ” triplestores, together with what has become the de-facto approach of converting NL queries to SPARQL queries before being evaluated, hinders the development of expressive NL query processors. The reason is that entity-based triples are not conducive to the development of semantic theories of complex prepositional phrases, and the development of such theories is made considerably more complex when translation to SPARQL has to be taken into account. An alternative approach, which uses “event-based ” triplestores, treats (bracketed) English queries as expressions of the lambda calculus which can be evaluated directly with respect to the triplestore. This approach facilitates the development of a formal denotational semantics of English queries which easily accommodates complex prepositional phrases. The approach described here could be used to develop a denotational semantics for a highly-expressive NL query language, and then that semantics could be used to guide the design of an NL query to SPARQL translator, thereby taking advantage of SPARQL optimizations.
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 | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.002 | 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".