GeoStartup Analytics
This project analyzes startup valuation data using basic regression models and interactive geospatial visualization. The workflow includes cleaning numeric fields, estimating valuation using linear regression and random forest models, computing feature importances, and evaluating predictive accuracy with R² and RMSE. The dataset is also mapped using Folium: countries are assigned by region, coordinates are merged, and startups are displayed on an interactive world map with marker clustering and a geographic heatmap. Circle marker sizes scale with valuation, and industries are color-coded to show geographic and sector patterns.