sentiment-analysis

What problem does it solve?

Extract and quantify market sentiment signals from unstructured text (news, social media, forums) to produce an objective, normalized score and explainable drivers for a specific asset or sector.

Core Features & Use Cases

• Multi-source aggregation: Combine news, social media, and forum signals with configurable conceptual weights to balance reliability and speed.
• Entity extraction and classification: Identify mentioned assets and sectors, classify text as positive/neutral/negative, and normalize scores to a 0.0–1.0 range.
• Trend detection and attribution: Compare current sentiment to prior windows to report Rising/Falling/Stable trends and list key drivers by source.
• Use cases: Generate a market sentiment brief for a ticker, feed sentiment signals into watchlists or reports, or provide per-source breakdowns for downstream skills.

Quick Start

Use the sentiment-analysis skill to produce a 24h sentiment report for NVDA combining news and social sources and return a structured sentiment report with score, trend, and per-source breakdown.