First you have to download all the data,

$ ./getdata.sh

And then go grab a coffee... and then probably lunch.

Now you're ready to start playing with data! The main file to be using is associate_contributions.py. It will load up all the contribution and votes data and join them together. Right now it limits itself to only votes that have happened in 2016 so that the load time is quick, but this can be changed by modifying the start_year parameter.

• Read in contribution data (contributions.py)
• Read in legislator metadata (legistlators.py)
  • Code written and it works, but numbers seem a bit lot... why??
• Read in votes data (votes.py)
• Correlate honorees in contributions to legislators in voting data
  • Votes data uses the bioguide id which we index on in legistlators.py
  • We'll need to do fuzzy searching... maybe remove common phrases (senator, sen., sir, ms, etc..) and then do best edit distance?

This will store my consulting project code for Insight Data Science Fellowship.

• Read in voting, bills, and contributions data.
• Group like bill subjects.
• Group like contributions.
• Train NN to predict bill voting outcomes.
• Create web application.