Chicago crime predictive model
Jupyter Notebook HTML Python
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
exploration
screenshots
templates
.gitignore
README.md
requirements.txt
webapp.py

README.md

EdPredPol

Chicago crime predictive model

This project has been developed for the Smart Data Hack 2016 at the University of Edinburgh, 15-19 February 2016.

Team: Adam Golinski, Lorenzo Martinico, Branislav Pilnan and Ondrej Bohdal.

Setup

We suggest using virtualenv. To install all the necessary dependencies run pip install -r requirements.txt.

Crime data can be obtained in the form of a CSV file from the City of Chicago data portal.
Weather data can be obtained from the NOAA website.

Structure of the project

We collaborated on data exploration and investigation using jupyter notebook and the semi human readable results with accompanying comments are available as .ipynb files in the exploration folder.

We've also created a simple Python Flask-backend webapp for web presentation of part of the work done and results obtained. You can run the webapp by running python webapp.py.

Mathematical model

We've used a Generalized Linear Model using statsmodel.api.GLM with Poisson distribution and log link.

Contact

Feel free to reach out to any of us concerning this project or anything else, every interest is most welcome :)