News for hackers. Stuff that matters.

A comment on Latent Dirichlet Allocation Surprisingly Well Correlated w/ Google Rankings. Parent comment

Re:Latent Dirichlet Allocation Surprisingly Well Correlated ... by nl on Tuesday September 07, 2010 @07:01AM (#1668562)
(Thanks for the reply)

Did you test non-LDA methods? Because to me it looked like a correlation between a set of related words and ranking for a topic related to those words.

Without testing non LDA methods I can't see what you've proved.

[ Parent | on HN ]
Re:Latent Dirichlet Allocation Surprisingly Well Correlated ... by will_critchlow on Tuesday September 07, 2010 @07:39AM (#1668614)
Yes. The chart in the post shows how low the correlation is for tf-idf. I believe the original also showed similarly poor results for LSI etc.
[ Parent | on HN ]