Finding Similar Political Interests


ThePublicWhip have parsed all Hansard data to get voting data. What would happen if this was combined with machine learning techniques (at the moment, just logistic regression) to try and map sets of votes to a party.

At the moment just used for recommending which party a member of the public should join, but there are lots of options for taking this forward.

Ideas for taking this project forward

Use more sophisticated ML techniques (some basic neural network back propagation would be the ideal next step). Also, use the data for other reasons.

Estimated costs for taking this project forward

Very cheap. Cloud computing (EC2) means extreme processing power can be rented for next to nothing.

About the data used for this project

Uses the Hansard-parsed voting data from ThePublicWhip. Pretty great data; no problems with it.

Project URL:

Twitter: @harryrickards

Created at

Rewired State: Parliament

Flickr Photos

Add photos by tagging them with rewiredstate:project="finding-similar-political-interests"