Triple
T1282467
| Position | Surface form | Disambiguated ID | Type / Status |
|---|---|---|---|
| Subject | Dick Costolo |
E27356
|
entity |
| Predicate | employer |
P7
|
FINISHED |
| Object |
E3345
|
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: Twitter | Statement: [Dick Costolo, employer, Twitter]
NED1
Entity disambiguation (via context triple)
gpt-5-mini-2025-08-07
Target entity: Twitter Context triple: [Dick Costolo, employer, Twitter]
-
A.
Twitter, Inc.
chosen
Twitter, Inc. was a major social media and microblogging company best known for its real-time short-message platform that shaped online news, politics, and public discourse worldwide.
-
B.
Instagram
Instagram is a popular photo and video sharing social media platform known for its visual content, stories, and influencer culture.
-
C.
Facebook
Facebook is a major global social networking platform that allows users to connect, share content, and communicate online.
-
D.
Tumblr
Tumblr is a microblogging and social networking platform known for its highly customizable blogs, fandom communities, and viral multimedia content.
-
E.
Reddit
Reddit is a large, community-driven social media and discussion platform organized into topic-based forums called subreddits.
- 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_69a496d3710c8190955dee8bc0dacb50 |
completed | March 1, 2026, 7:43 p.m. |
| NER | Named-entity recognition | batch_69a4c0b47be08190828a1c0a11d94ce8 |
completed | March 1, 2026, 10:41 p.m. |
| NED1 | Entity disambiguation (via context triple) | batch_69acb2fd091081908f4a60d7ce300c90 |
completed | March 7, 2026, 11:21 p.m. |
Created at: March 1, 2026, 7:50 p.m.