Predicting Low Income Low Access Tracts in US

Oct 30, 2022

Summary of the project:

  • Provided Exploratory data analysis using Choropleth maps to identify regions with high food insecurity

  • Conducted Linear Regression Analysis and Linear Discriminant Analysis on relevant metrics

  • Used K-Nearest Neighbor Classifier using relevant variables to classify Low Access regions

  • Provided a Logistic Regression model to predict Low Income-Low Access tracts based on other relevant variables with a high AUROC ( Area under the Receiver Operating Characteristics curve) of 0.89

You can find the slides for the presentation of this project here and the corresponding report here .

ML Statistics
Rohan Shinde
Rohan Shinde
Risk Analyst