I'd like to perform semi-supervised LDA (Latent Dirichlet Allocation) in the following sense: I have several topics that I'd like to use, and have seed documents that relate to these topics. I'd like to run LDA to classify other documents, and potentially discover other topics.

I would guess there is work done on that, as the problem is natural, and the LDA framework seems to suggest it, nevertheless, I'm not an expert and do not know about such work. Can you guide me to papers or tools ?

  • 1
    $\begingroup$ One example is L. Fei-Fei and P. Perona. A bayesian hierarchical model for learning natural scene categories. In CVPR, 2005. But your question is unclear. Do you mean that for each document there is a single topic? Do you mean a topic corresponds to a group of documents (a class)? Or more than one topic? How do you plan to do classification? What other topics would you need? $\endgroup$ – SheldonCooper Mar 13 '11 at 8:08

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.