Data Science2021Data Scientist
East-Central London & Birmingham Venues
IBM Data Science Professional Certificate capstone
Compared East-Central London and Birmingham by venue category using Foursquare, geolocation, Folium maps, and K-means clustering (6 clusters in ECL, 2 in Birmingham).
The Problem & Engineering Constraint
The Core Challenge
Neighbourhood comparison is noisy if you only look at raw venue counts. The brief needed cleaned data, maps, and unsupervised structure.
Technical Architecture & Approach
Engineering Solution & Implementation
Pandas cleaning, Matplotlib visualisation, Folium + Foursquare + Geolocator APIs, then scikit-learn K-means. Discussed how venue-category clusters apply to each city.
Measured Production Impact
Verified Outcomes & Deliverables
Formed 6 venue clusters in East-Central London and 2 in Birmingham.
Capstone for the IBM Data Science Professional Certificate.
Technologies & Components
System Tooling & Technologies
PythonPandasFoliumFoursquare APIGeolocatorScikit-LearnK-means