notre livre

Define your comfort of life with the data

This research and development work aims to create the first well-being score in the city based on the available open-data and private data as well as the related human factors to the comfort of life.

As part of the Aquitaine Region’s support for technology transfer, 10h11 had the opportunity to work with CATIE and the ENSC engineering school to propose an innovative approach to the notion of living comfort through data, based on a scientific approach to define these indicators and automatically collecting information on the Internet to make the algorithms work.

This research and development work aims to create the first well-being score in the city based on the available open-data and private data as well as the related human factors to the comfort of life.

In concrete terms, the work was divided into several phases :

• Establish a list of data sources and assign each one a reliability rating;

• Implement the logic of retrieving this information and storing it in BigData;

• Study and define, based on cognitive studies, what are the criteria that make it possible to define a quality of life comfort score;

• Implement these information extraction algorithms to display a note per neighbourhood;

• Propose an adapted visualization of the prototype.

Focus on data structuring

We have carried out work to consolidate the database, but also to structure it in order to be able to weight all the types of data according to the definition of living comfort expressed by the panel.

Consider a criterion A, for example “Shops and restaurants”: we can see it as a superset of A.1 and A.2, for example “Bars and restaurants” and “Local shops”.

We therefore imagined deploying all our criteria in this tree form, with several levels, where the sub-criteria would be the leaves.

This structure allows us to fully exploit the relevant data and to prioritize them according to their importance for the panel.

Focus on the algorithm 

Once our database was structured, we had to be able to rank a number of alternatives in order of choice on the basis of a set of favourable or unfavourable criteria.

The solution we have chosen is called Topsis (Technique for Order of Preference by Similarity to Ideal Solution).

Indeed, TOPSIS allows us to determine, for each alternative, a coefficient based on the distances (Euclidean) between each alternative on the one hand and the ideal favourable and unfavourable solutions on the other hand

NB: an alternative is said to be favourable if it is furthest from the worst alternative and closest to the best alternative; on the contrary, an alternative is said to be unfavourable if it is closest to the worst alternative and furthest from the best alternative)

In addition, TOPSIS has the ability to provide us with standardised results, fairly simple to implement, with a relative ranking of the criteria (it makes it possible to order the criteria in order of preference, one against the other) and to provide a calculation based on the average distance of the establishments in a circular area, weighted by the number of establishments in the calculation area (this makes it possible to treat the problem from a point located at the border of a district).

These criteria make it possible to respond to the cartographic problem that we were considering on the comfort of life in order to visualize the comfort scores according to the different districts of the same city.

Focus on the mesh size and the minimum calculation area

We started from the premise that the user is able to have preferences on the different districts or areas of the city. In this case, the application should allow him to compare areas on several levels of detail. Thus we had to aggregate the detailed data in order to provide a satisfactory experience (performance).

To meet this meshing requirement, the map with regular hexagons is the most optimal form (the calculation of establishments is carried out in a circle of radius of a grid, automatically generated from a single point (centre of France), thus making it possible to grid the whole of France.

For each zoom level, the living comfort score is recalculated according to the criteria recorded in the area. For the first prototype, only one zoom level has been set up.

Focus on data visualization

Our prototype includes MapBox services allowing us to work on a refined mapping, in grey areas, to highlight the scores of each area.

The purple color materializes the scoring areas and the opacity of the color determines the intensity of the score of each hexagon. Finally, the open-data and private data sources are structured on the map by means of yellow, orange, green and blue dots. Each colour can be assimilated to a category of data.

When loading a loader comes to warn the user of the work of our server to process the request and display the information.

5 click areas are available:

Each corner of the screen contains a clickable element to return to home, expand the menu, zoom in or get more information.

The hexagons of the mapping are clickable in order to provide the information related to the data that allowed to determine the comfort score of the area concerned. Whenever you click on a hexagon, the map automatically repositions itself to center the interface on the clicked area. This option allows us to make it easier for the user to navigate and understand the information.


To discover the prototype, the scores and navigate on the map, here is the test URL at your disposal:

Any feedback is welcome, do not hesitate to contact our team on this subject.

Social Media Auto Publish Powered By :
Cart Overview