DNP-XGboostPitalito

DNP-XGboostPitalito contains building footprint layers estimated from an XGBoost classification model applied to satellite imagery for the Municipality of Pitalito. The model was developed during the DataSandbox Colombia 2020 by the Data Science Unit of the National Planning Department (DNP). The dataset was last updated on May 18, 2026.

Use Cases

• Classifying building presence from satellite imagery using the probability column as a model confidence score.
• Aggregating building coverage by administrative area using the municipio and polygon columns for urban planning.
• Validating or comparing automated footprint extraction models against this XGBoost-derived baseline using the provided polygon geometries.
• Creating visualizations of estimated construction density across Pitalito by mapping the probability values associated with each polygon.

Strengths

• Derived from a documented machine learning methodology (XGBoost) applied to satellite imagery.
• Includes a probability column, which provides a confidence score for each classification.
• Available in multiple open formats, including CSV, JSON, XML, and RDF.

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download.
• Row count and dataset size are unknown, which may limit suitability assessment.
• The license and specific author are not provided in the available metadata.

Provenance

Source

Data Science Unit of the National Planning Department (DNP), Colombia.

Collection Method

Estimated from satellite imagery using an XGBoost classification model developed during DataSandbox Colombia 2020.

Freshness

Last updated 2026-05-18 18:47:57; freshness should be verified.

Geography

Municipality of Pitalito, Colombia. Projection: MagnaSirgas 3115.