🚀 Topic Trend Aggregator

Turn raw news noise into actionable strategic intelligence. Identify emerging trends, track sentiment momentum, and generate high-level strategic narratives automatically.

---

What is Topic Trend Aggregator?

The Topic Trend Aggregator is the intelligent engine room of the Visita Content Intelligence Suite. It is an AI-powered tool that ingests hundreds of raw news articles from multiple sources, uses GPT-4o to cluster them into coherent topics, and scores them based on momentum and sentiment.

Unlike simple news scrapers that just give you a list of links, this Actor reads and reasons about the news for you. It acts as a 24/7 dedicated analyst that tells you what is happening, why it matters, and how people feel about it.

Why use Topic Trend Aggregator?

• Detect Emerging Trends Early: Spot viral stories and market shifts before they become mainstream news.
• Automate Market Research: Save dozens of hours of reading by letting AI synthesize hundreds of articles into clear summaries.
• Strategic Intelligence: Move beyond daily noise with our "Meta-Analysis" mode that identifies long-term macro-narratives.
• Seamless Integration: Automatically feeds clean, scored data into the Sentiment Compass for actionable insights.

---

🌟 Key Features

• 🤖 AI-Powered Clustering: Uses OpenAI (GPT-4o) to read headlines and group related stories into single, coherent "Trends" to eliminate duplicates and noise.
• 📉 Noise Reduction: Filters out isolated, low-impact stories to focus only on topics with actual momentum.
• 🧠 Two Operation Modes:
  • STANDARD: Daily scouting for fresh trends (Micro-Trends).
  • META-ANALYSIS: Monthly strategic roll-up of all past trends into "Macro Narratives."
• 🛡️ Cost Safeguards: Built-in limits (max 200 articles per source) prevent runaway API costs while ensuring a high signal-to-noise ratio.
• 🧠 Persistent Memory: Remembers every URL it has ever seen to ensure you never pay to process the same news twice.

---

🔌 The Visita Content Pipeline

This Actor is designed to be Step 1 in a powerful content automation pipeline. It integrates seamlessly with our other tools to go from Raw Data to Published Content.

graph LR
    A[News Sources] -->|Raw Data| B(Topic Trend Aggregator)
    B -->|Clean Trends Dataset| C{Sentiment Compass}
    C -->|Opportunity Scores| D[Content Blueprint AI]
    D -->|Final Articles| E[Publish]

Step 1: Topic Trend Aggregator (This Actor)

• Input: Select "World News Intelligence", "Tech & AI", or other sources.
• Action: Scrapes, clusters, and identifies "What is happening right now."
• Output: A clean JSON dataset of emerging trends.

Step 2: Sentiment Compass (Integration)

• Action: Run the Sentiment Compass actor.
• Input: Select the Dataset ID produced by this aggregator.
• Result: It will analyze the emotional valence (Positive/Negative) and assign an "Opportunity Score" to each trend, helping you decide which topics are worth covering.

Step 3: Content Blueprint AI (Creation)

• Action: Use the high-scoring trends from the Compass.
• Input: Feed the trend data into the Content Blueprint AI.
• Result: Automatically generate blog posts, social media scripts, or strategic reports based on the verified intelligence.

---

💰 Pricing

This Actor uses a transparent Pay-Per-Event model. You only pay for value generated (trends found), not for the time it runs or the empty searches.

Event	Price	Description
Trend Detected	$0.015	Charged for each distinct micro-trend discovered (e.g., "Nvidia Stock Surge").
Macro Narrative	$0.15	Charged for each high-level strategic narrative generated during Meta-Analysis (e.g., "The Shift in Global AI Chip Supply").

(Note: Pricing event names may appear as TREND_GENERATED or trend-detected in your invoice depending on the active platform version).

How much will it cost to analyze trends?

• Daily Run: If the actor finds 20 distinct trends from 15 news sources, you pay: 20 * $0.015 = $0.30.
• Monthly Report: If you run a Meta-Analysis that generates 5 Strategic Narratives, you pay: 5 * $0.15 = $0.75.

⚠️ Important Note on OpenAI Costs: This Actor requires your own OPENAI_API_KEY. To protect you from excessive OpenAI bills, this actor strictly limits processing to the top 200 articles per source per run. This ensures you get the highest quality "signal" without paying to process "noise."

---

⚙️ How to Use

1. Choose your Operation Mode

• 🚀 STANDARD (Daily): Select this to fetch fresh news right now. Ideal for running every morning to see "what's new."
• 🧠 META_ANALYSIS (Monthly): Select this to look back at the last 30 days of trends and ask the AI to write a "Big Picture" strategic report.

2. Select News Sources

Pick from our curated list of intelligence streams. You can select multiple sources at once:

• World News Intelligence
• Finance & Global Markets
• Tech & AI News
• Retail & Ecommerce Innovation

3. Set Execution Strategy

• LIVE: Triggers a fresh scrape of the news sites. This is slower but guarantees the most up-to-date information.
• ARCHIVE: Uses the most recent cache from our database. This is instant and incurs zero scraping cost—perfect for re-analyzing data you just collected.

4. Start & Connect

Click Start! The actor will output a clean JSON dataset. Copy the Dataset ID and paste it into the Sentiment Compass to continue your workflow.

---

Output Data

The Actor produces a structured JSON dataset containing clustered trends. Each item represents a single trend, not a raw article.

Example Output:

{
  "cluster_id": "World News-1732085383",
  "cluster_topic": "AI and Tech Independence in Europe",
  "summary": "Europe is striving for AI independence to avoid becoming a technological colony of the US or China.",
  "internal_trend_score": 3.55,
  "sentiment_score": 0.18,
  "sentiment_label": "POSITIVE",
  "articles_count": 3,
  "mentioned_organizations": ["Nvidia", "European Commission"],
  "articles": [ ... list of source articles ... ]
}

---

FAQ & Support

Can I use this without the Sentiment Compass?

Yes! The output dataset is standard JSON. You can download it and use it in Excel, Google Sheets, or your own custom Python scripts for market research.

Why are there multiple datasets in my run?

The actor orchestrates other scrapers (like World News Intelligence) to get data. These scrapers produce their own "raw" datasets (e.g., world-news-200-items). You can ignore these. Only use the "Default" dataset, which contains the final, analyzed trends.

Is it legal to scrape news data?

Our scrapers collect publicly available headlines and summaries. We do not bypass paywalls or scrape private user data. However, you should always ensure your use of the data complies with copyright laws and fair use policies in your jurisdiction.

---

🔗 Related Actors

Build your complete intelligence pipeline with the Visita Suite:

• Sentiment Compass - Analyze the emotional impact of your trends.
• World News Intelligence - The raw data engine for global news.
• Content Blueprint AI - Turn trends into content automatically.