M.S. Applied Data Science - Capstone Chronicles 2025

14

Figure 5 Top 5 Business Categories

Degree Distribution. The degree distribution describes how widely connected each business is to other businesses. Our analysis was performed on a networkx directed graph produced where a node may have both in and out connections with other nodes. Examining the degree distributions allowed us to understand how connectivity within the network was spread across businesses or in many cases, concentrated among a small number of highly connected nodes which we refer to as “hubs.” Across all views of our network, on in-degree distribution we see spikes around 5,000– 5,500 in-degree. This indicated a small set of businesses that attract customers from many other businesses.These are likely the retailer hubs (e.g. Walmart, Target), high traffic chains (e.g. Starbucks, McDonalds) and “essential” locations (e.g. gas stations). For physical, in-store locations, hubs include 7-Eleven, Costco, CVS, McDonalds, and Chevron. The peak near low degree nodes indicated the presence of local or niche business with little overlap. The central peaks suggested a grouping of similar attraction levels. Ultimately, this pattern supported the

222