Latent Semantic Analysis/Indexing Library for C++

Question

Is there a C++ library for LSA/LSI? Preferably MIT, BSD, Apache,... license - no GPL.

Answer 1

There are several C/C++ implementations of Latent Diriclet Allocation (LDA), Latent Semantic Analysis (LSA) and Topic Modelling listed on David Blei's page. http://www.cs.columbia.edu/~blei/topicmodeling.html and on Berry and Dumai's page http://web.eecs.utk.edu/research/lsi/

Here's a short-list of C/C++ implementations:

• lda-c : D. Blei's implementation with variational inference for LDA.
• class-slda : C. Wang's implementation of supervised topic models with a categorical response.
• hdp: C. Wang's implementation of Hierarchical Dirichlet processes
• Gibbs Sampling: Phan and Nguyen's implementation of LDA using Gibbs Sampling -PHP LSA: http://sourceforge.net/projects/phplsa/
• lsa_prj: https://github.com/matpalm/lsa/tree/master/lsa_prj

If you like a python version with tutorial: see http://www.puffinwarellc.com/index.php/news-and-articles/articles/33-latent-semantic-analysis-tutorial.html?showall=1