We mapped the French cultural asset

1
Overview of the project map


I have been working for a government startup, as a Full Stack developer, on a project designed to map all the cultural assets of France. In this blog, I will explain the project and you will be able to see the demo below.

What is it about

More than 800K assets were referenced and geolocalized since several years across multiple databases including national heritage monuments, museum’s artworks, and other protected historic treasures. You can find entities such as churches, archaeological sites, and castles of unique cultural and historical significance.

As the data started to be collected several years ago, the quality and the format was too diversified to be readily used. Many different systems of projection could be find such as Lambert 93, Lambert2, WGS84 or Gauss

Local to Global knowledge

“Making the most important works of France accessible to all” is, in essence, the mission of the Ministry of Culture.

One of the ways to serve this mission is to improve and upgrade tools used by the government and museums in order to inventorize the rich cultural assets on the French territory.

The geolocalized data could serve as an important source of cultural assets for many sectors such as education or tourism. It’s also important for the local history sharing as the community and contribution dimension of the project is significant

Church from the 11th Century

Technically speaking

The map layer used is provided by mapbox which use openstreetmap data. Mapbox provides a the easiest way to display map and create map animations in JavaScript.

The data layer is using the search engine Elasticsearch. Elastic is an open source search engine, initially focused on text search but also providing some location capabilities.

Elasticsearch manages hashmap indexing of data which gives a first raw clustering system

Finally, a custom clustering has to be done on top of the Elasticsearch hashmap system. It has been esthetically improved using a weighted K-mean cluster algorithm.

Culture is everywhere

Having cultural assets inventoried, located, classified, documented and openly shared to everyone enables culture valorization and opening of new opportunities. This is a reminder that culture is everywhere if you know where to look.

Elements