> For the complete documentation index, see [llms.txt](https://intelfactor.gitbook.io/docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://intelfactor.gitbook.io/docs/home/edge-pipeline/data-model-causal-triples.md).

# Data model: causal triples

The learning unit that connects defects to verified fixes.

IntelFactor’s core learning unit is a **defect → cause → outcome** triple.

### Example fields

* `defect_id`.
* defect type, severity, confidence.
* correlated parameter drift (JSON).
* cause hypothesis + cause confidence.
* bilingual explanation (when enabled).
* SOP-linked recommendation.
* operator action: accepted / rejected / modified.
* measured outcome: defect rate after, time to baseline.
* status: pending / verified / disputed.

### Why this matters

Detection datasets answer: “what does a defect look like?”

Triples answer: “what caused it, and what fixed it?”