For this project, I chose a dataset from OpenDataPhilly focusing on the food landscape in Philadelphia neighborhoods. The dataset ‘Neighborhood Food Retail’ examines the availability of stores that offer healthy food options such as supermarkets and farmers’ markets, versus those offering fewer healthy options, like dollar stores and convenience stores. The outcome variable, “ACCESS,” aims to predict whether an area has access to healthy food options. Initially, the “ACCESS” column held three categories: being “Low Access,” “High or Moderate Access,” or “No Access” to healthy food options. To make it compatible with the decision tree model, I changed cells holding “Low” or “No access” to 0 and “High” or “Moderate” to 1. Utilizing Python and the CSV file, I was able to create a decision tree to determine the traits that determine the likelihood of a neighborhood having low quality options. Analyzing this dataset provides valuable insights into the food landscape and access to healthy food options in Philadelphia neighborhoods
The data suggests that supermarkets may not see value in opening locations in these neighborhoods, possibly due to the residents’ limited ability to afford fresh produce and higher quality products. This information can be useful for charities, funding organizations, and initiatives like the Cherry Pantry, as it highlights the specific areas that need help.
Decision Tree Results and Code
