Disaster aid request vs fraud-pattern text cues
Judges: one aid application narrative. Questions: noul:template_text - boilerplate seen across apps?; noul:internally_inconsistent - .... Action: Add cue to a fraud score; manual review for flagged.
AA clean typed decision (yes/no, a pick from a list, or a level on a scale) that is concrete, repeats routinely, and has meaningful value.4.73
Key facts
- Vertical
- Government & nonprofit
- Function
- Legal & finance
- Status
- Idea
- Volume
- routine
- Value
- major
- Risk
- high
- Evidence
- —
- Flags
- human-review
Build this with a classifier
Define a typed decision with a bounded answer, then evaluate it on examples.
{
"decision_type": "yes_no",
"question": "Does this input match the decision in “Disaster aid request vs fraud-pattern text cues”?",
"input": "<input to classify>",
"output": "yes | no"
}Related use cases
Cite this
Copy a link in your preferred format.