Triple
T1181110
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Clarice Starling |
E25138
|
entity |
| Predicate | creator |
P184
|
FINISHED |
| Object | Thomas Harris |
E125454
|
NE FINISHED |
How this triple was built (2 steps)
Every LLM step that produced this triple, in pipeline order — named-entity classification, the disambiguation choices (the exact options shown, with the pick highlighted), and the generated description. The batch + timestamp of each is in the Provenance table below.
NER
Named-entity recognition
gpt-5-mini
Instruction
Given a phrase, classify it is english named entity (e.g., persons, organizations, works of art) in Latin script, or not (e.g., literals, dates, URLs, verbose phrases). For disambiguation, the statement where the phrase occurs as object is also given. Please return a JSON object with `phrase` (string, the phrase being analyzed) and `is_ne` (boolean, indicating whether the phrase is a Named Entity).
Input
Phrase: Thomas Harris | Statement: [Clarice Starling, creator, Thomas Harris]
NED1
Entity disambiguation (via context triple)
gpt-5-mini-2025-08-07
Target entity: Thomas Harris Context triple: [Clarice Starling, creator, Thomas Harris]
-
A.
Thomas Harris
chosen
Thomas Harris is an American novelist best known for creating the character Hannibal Lecter in his series of psychological horror thrillers.
-
B.
Michael Connelly
Michael Connelly is a bestselling American crime fiction author best known for his Harry Bosch and Lincoln Lawyer series.
-
C.
Lee Child
Lee Child is a British thriller author best known for creating the Jack Reacher series of novels.
-
D.
James Herbert
James Herbert is a film editor known for his work on major feature films, including the fantasy action movie "King Arthur: Legend of the Sword."
-
E.
Tess Gerritsen
Tess Gerritsen is an American physician-turned-bestselling author best known for her crime thriller novels, particularly the series featuring detective Jane Rizzoli and medical examiner Maura Isles.
- F. None of above.
- G. Unsure - the case is ambiguous/there is not enough information to decide.
Provenance (3 batches)
The batch behind each pipeline step, in order, with when it ran. Timestamps are batch-level — stages were processed in waves, so the object chain (NER → NED1 → NEDg → NED2) reads in order, but predicate / elicitation batches can sit in a different wave.
| Step | Stage | Batch ID | Status | When |
|---|---|---|---|---|
| creating | Elicitation | batch_69a494267b4c819088c97a59182bf56a |
completed | March 1, 2026, 7:31 p.m. |
| NER | Named-entity recognition | batch_69a4bd32c5f48190b4e2d39fa052cbb7 |
completed | March 1, 2026, 10:26 p.m. |
| NED1 | Entity disambiguation (via context triple) | batch_69ac6f1f1c188190a96f5718c4e7d59d |
completed | March 7, 2026, 6:31 p.m. |
Created at: March 1, 2026, 7:45 p.m.