How It Works
Three Steps to Prediction
Our ML pipeline transforms raw data into actionable insights
1
Input
Provide stock symbol and date range
Historical price data from Yahoo Finance
Technical indicators (MA, volatility)
Configurable time windows
2
Sentiment Fusion
Multi-source sentiment aggregation
News sentiment via VADER analysis
Google Trends search interest
Wikipedia pageview analytics
3
Prediction
ML-powered next-day forecast
Random Forest regression model
Explainable feature breakdown
Exportable CSV predictions
Developer-First
Built for Developers
Production-ready ML pipeline with full transparency and reproducibility
Live Data Sources
Real-time data fetching from Yahoo Finance, Google News RSS, Google Trends API, and Wikipedia pageviews. No stale datasets.
yfinancepytrendsfeedparser
Explainability First
Every prediction includes sentiment breakdown, feature importance, and validation metrics. Understand exactly how the model works.
Sentiment ScoresFeature Logs
Reproducible Results
Fixed random seeds, versioned dependencies, and deterministic training. Run the same experiment twice, get identical results.
random_state=42requirements.txt
CSV Export Ready
Export features and predictions as CSV for further analysis, visualization, or integration with your own tools and workflows.
features.csvpredictions.csv
Powered By Industry-Leading Technologies
