Predicting crime types in the City of Boston
Summary
This website contains the analysis and predictive results for a project exploring the prediction of crime types in the City of Boston, based on a number of different features.
An accompanying repository containing the Python code used to perform this analysis and associated findings can be found here: https://github.com/sedelmeyer/predicting-crime. Please look at it extensively as it contains all of our work.
This website was built as a part of the requirements for the CS109a class at Harvard “Introduction to Data Science”.
Research questions
Given the (1) variety preliminary datasets we were able to find, (2) the results of our initial EDA on a subset of those datasets, and (3) the number of additional predictors we currently have under development, we have decided to define our project questions in fairly broad terms similar to how they were proposed in the original project description:
-
Given environmental and demographic features of a specific location in the City of Boston, can we predict (from a defined subset of crime types), which type of crime is most likely to occur at that location?
-
What effects do specific location-based features have on the type of crime most likely to occur at that location?