Predictive Analytics for SF 311 IPython Notebook

[Ted Fujimoto]

Hi all,

Instead of an app, I was thinking about an IPython notebook on how to best predict service requests from SF 311 datasets. DSSG and Kaggle have already done similar projects with Chicago's 311 service requests. Right now, the plan is to use Python data tools (statsmodels, scikit-learn, pandas, etc.) to see how generalizable other methods are and make any necessary improvements. The hope is to post our code, methodology, results, and advice on an IPython notebook so that others can use our techniques to improve their own cities.

If you want to do data science, machine learning, statistics or data visualization for good, this project is for you! Please respond if you are interested and want other relevant materials to prepare for this project.