- Then I made the words of each topic into a word cloud, because it’s still 2008 and word clouds are cool.
- I used Latent Dirichlet Allocation to build 20 sets of words that each match with a “topic” in the text* of the budget.
- I topic modeled all the Estimates of Appropriations documents from the NZ 2016 budget.
- Thanks to @garibaldu for the LDA code.
Read the full article, click here.
@droneale: “I topic modelled #Budget2016, because why not mindlessly apply machine learning to govt docs”
Word clouds of the 20 topics from topic modeling the NZ 2016 bidget with LDA
Topic modeling Budget 2016