2019-06-26, 11:30–12:00, Room 1
Geopandas and CARTO provide valuable tools to analyze multivariate geolocated data, enabling data-driven decisions in the context of site planning.
Spatial data from a variety of sources are increasingly used to target marketing campaigns for an optimal audience. In this talk, I will show how Geopandas and CARTO can be used in site planning applications to combine different data sources (e.g. mobile, financial, demographic and point of interest data) to find locations that show similar characteristics. In order to compute a similarity score with respect to an existing site, the method uses probabilistic principal components analysis, therefore properly accounting for correlation in the data as well as for missing values, resulting for example from anonymization algorithms. Results are finally visualized in a vector map, allowing the user to easily derive data-driven insights when opening, relocating or consolidating location sites.