Perform either a k-means analysis or a hierarchical clustering analysis to compare any selected list of neighborhoods in your city. ** You may find this task far more manageable if you reduce the number of neighborhoods that you're dealing with. The results will be more meaningful that way, too. ** Of any section of the project, this one offers the most opportunity to be creative and take risks. Think about feature engineering, too -- how/when/where can you create new variables based on existing ones? ** group_by() and summarize() from the tidyverse can be quite helpful here (or, if you prefer to use aggregate() and tapply(), those can help to accomplish a similar task