Latent Dirichlet Allocation on Tweets

Re:Latent Dirichlet Allocation on Tweets by dhotson on Sunday September 07, 2014 @08:42PM (#8282955)

Very cool!

I've read that LDA doesn't work well on short documents. Your approach of concatenating all tweets for a user appears to work quite well. One other technique I've seen is to concatenate multiple tweets together that contain the same hashtag.

One of our intern students at 99designs did some work on applying LDA to classify graphic design tasks:

http://99designs.com.au/tech-blog/blog/2014/01/22/Swiftly-Ma...

.. you might find it interesting. :)

[ Parent | on HN ]

Re:Latent Dirichlet Allocation on Tweets by biomimic on Sunday September 07, 2014 @10:34PM (#8283190)

Also used to classify genes related to lifespan - LDA on Statistical modeling of biomedical corpora: mining the Caenorhabditis Genetic Center Bibliography for genes related to life span. http://www.ncbi.nlm.nih.gov/pubmed/16681860

[ Parent | on HN ]