This map is a sample for now. A polygon is an area on the map. Every area of the dataset is included, with at most 100 random polygons per area.

A model read each sentence of the polygon text. The text comes from OpenStreetMap description tags. The model gave each sentence a label:

• yes: the sentence is about land use.
• no: the sentence is not about land use.
• failed: the model answer cannot be read.
• not split: the source did not split the text into sentences.

Hover over a polygon to read its text. Use the slider to hide small polygons.

Data: osm-polygon-description-tag-landuse (ODbL). Code: GitHub.