Term-extraction algorithms have the disturbing property of making things sound more interesting than they really are. These are WordPress’s suggested tags for the last blog post:

community organisers, multilateral agreement on investment, plymouth argyle, james purnell, crash investigation

Way more fun than just another future of the Labour Party post. This is of course inherent in the methodology. You’re trying to identify statistically significant variance in the corpus used, so the weirdness pops out from the background. Like teal and orange.


  1. I’ve been amused watching the Facebook software try to successfully suggest adding friends to my categories — distant cousins (3d or 4th), people I knew before 1984, lawyers who work on Guantanamo Bay cases. These are geographically diverse groups, and my criterion for categorization of relationship to them is fairly distinct from that which applies to their relationships with their neighborhood friends, spouses, in-laws, etc.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: