M.S. Applied Data Science - Capstone Chronicles 2025

10

business. The three dimensions were “local,” filtered only to brands that matched transactions in the same zip code, “physical,” filtered to brands that customers spend money with in person, and “same category,” filtered to only brands in the same top category. Exploratory Data Analysis In this section we explore the composition and distribution of businesses across San Diego. We analyzed univariate features such as business count, customer count, spend distribution by city, as well as top business categories. Additionally, we explored the business graph’s degree connectivity Figure 1 SafeGraph Entity-Relationship Diagram

218