Project completed as Capstone Project for the IBM Data Science Professional Certification.

This analysis aims to support entrepreneurs by giving them a clearer understanding of the business environments for each district and neighborhood through clustering at the district and neighborhood levels according to the most popular business categories in each district. Cluster analysis of available data (described below) will allow entrepreneurs to select the appropriate location for their business to ensure it is suited to local tastes to increase chances of success.

Through combining Berlin district shapefiles with demographic data as well as business data retireved from the Foursquare Places API, a KMeans clustering model was developed to group districts and neighborhoods that are similar in terms of most commonly found business types.

Refer to the project notebook for the analysis with code. For the final deliverable, refer to either the final report or this blog post. Comments and feeback are welcome to this GitHub account, on LinkedIn or by email at [email protected].