|
@@ -5,7 +5,7 @@ Return STRICT JSON with EXACT keys: { topic, entities, sentiment, sentimentScore
|
|
|
FIELDS:
|
|
FIELDS:
|
|
|
- topic: one of [crypto, macro, regulation, ai, other]
|
|
- topic: one of [crypto, macro, regulation, ai, other]
|
|
|
- entities: named people, places, orgs, conflicts, and finance/crypto terms
|
|
- entities: named people, places, orgs, conflicts, and finance/crypto terms
|
|
|
- (BTC, ETH, ETF, SEC, ECB, Fed, euro, inflation, rates). Canonical forms. 1–5 words each.
|
|
|
|
|
|
|
+ (BTC, ETH, ETF, SEC, ECB, Fed, euro, inflation, rates). Canonical forms. 1–5 words each. Good: "USS Gerald Ford", Bad: "american aircraft carrier"
|
|
|
- sentiment: "positive" | "negative" | "neutral"
|
|
- sentiment: "positive" | "negative" | "neutral"
|
|
|
- sentimentScore: float –1.0 to 1.0, consistent with sentiment label
|
|
- sentimentScore: float –1.0 to 1.0, consistent with sentiment label
|
|
|
- keywords: 2–4 thematic tags, 1–2 words each. Noun phrases only (e.g. "drone strikes",
|
|
- keywords: 2–4 thematic tags, 1–2 words each. Noun phrases only (e.g. "drone strikes",
|
|
@@ -13,7 +13,7 @@ FIELDS:
|
|
|
|
|
|
|
|
TASKS:
|
|
TASKS:
|
|
|
1. Infer the best topic.
|
|
1. Infer the best topic.
|
|
|
-2. Extract all named entities from headline and summary.
|
|
|
|
|
|
|
+2. Extract all named entities from headline and summary. No general categories.
|
|
|
3. Assign sentiment from tone and wording.
|
|
3. Assign sentiment from tone and wording.
|
|
|
4. Choose keywords that capture themes, not entities.
|
|
4. Choose keywords that capture themes, not entities.
|
|
|
|
|
|
