Textual analysis of urban greening projects in the press

The non-profit portal GruenStattGrau is a knowledge-management platform focused on urban greening. With the consideration for climate change, water management and air pollution the competency center aims to promote the development of green rooftops, facades and urban farming.

The data4good Hackathon project will focus on a textual analysis of the relevant literature (magazines, websites, blogs, …), to understand the geographical distribution of urban greening projects, as well as the development of topics over time.


A team of very talented data scientists have decided to help solving this challenge. The main objective is to help GSG to analyze and extract insights from unstructured content.

The analysis will help to discover the obstacles the industry is struggling with. The results will enable GSG to detect potential areas for green roofs. Based on the detection they can suggest funding opportunities for greenery projects

Data will be extracted from magazine articles published in the last few years. The articles are provided in PDF format. Additional open datasets:

Sunset in the mountains
Sunset in the mountains
Sunset in the mountains


The following map represents the number of times that countries are mentioned in the texts from the Grünstattgrau. Note that the countries mentioned are not only specific to buildings with green roof tops. That distinction in the analysis has to be better worked out.

Due to the higher number of mentions in Germany, we created a separate map with data describing cities.

Download the results presentation

Download PDF