SentiLab

Data Science Lab for Market Sentiment

Predict asset prices based on sentiment analysis. Experiment with how news sentiment from specific sources, topics, or credibility levels affects price. Save your trading strategies permanently and verify them over time.

Start Experimenting

Three Experiment Types

Correlation Sweep

Find the optimal time lag where sentiment best predicts price movement. Discover which combination of news parameters most closely correlates with asset price changes.

r = 0.431, lag = +6h

Source Predictability

Rank news sources by their predictive power. Discover which sources consistently lead price movements (oracles) and which follow them (echo chambers).

45 sources ranked by r-value

Narrative Clustering

Detect when multiple sources converge on the same narrative. A "sell the news" signal — when everyone is saying the same thing, the move may already be priced in.

128 data points analyzed

How It Works

1

Choose an Asset

BTC, ETH, Gold, EUR/USD, and more

2

Configure Parameters

Time range, quality filters, lag window

3

Run Experiment

AI processes your request in seconds

4

Analyze & Save

Save strategies, publish, verify over time

What You Can Do

Save trading strategies permanently and verify them over time
Experiment with how sentiment from specific sources affects price
Filter experiments by credibility, quality, and market impact
Publish experiments as templates for the community
Compare predictive power of different news sources
Detect market narrative convergence signals

See What Others Have Discovered

Browse public experiments published by the community.

Browse Public Experiments →

Ready to run your first experiment?

Create a free account and start exploring the relationship between market sentiment and asset prices.

Create Free AccountSign In