What packages for pLSA (Probabilistic Latent Semantic Analysis) and LDA (Latent Dirichlet Allocation) do you recommend for Python? Java and C++ are OK too, but Python preferred. Your experiences are highly appreciated.
There are many libs, so I don't know which to choose.
Gensim seems to be most popular one. It's on PyPI so, you can just install it using
sudo easy_install gensim
There's a good LDA example implementation in python that can be found here: http://www.cs.princeton.edu/~blei/topicmodeling.html and the tarball link is here: http://www.cs.princeton.edu/~blei/downloads/onlineldavb.tar
It's not a library, but rather an example implementation that automatically downloads wikipedia articles and runs LDA. But I've found it to be extremely helpful as a reference for implementing my own LDA code.
