I am new to LDA topic modelling. I am using gensim and am able to generate topics that make sense. Using 25k of documents, I can also print them using print_topics. I am aware that we need use the generated model to classify any new document. I would like to know the topics for each training document. Do I have to treat them as new documents to find their topics?

lda.print_topics(num_topics=22, num_words=8)
  • $\begingroup$ Could you add some more code, i.e. the model you estimate. So people can follow your approach and make meaningful suggestions. Which platform are you on? Python? You could add a tag. $\endgroup$ – Peter May 31 '19 at 14:28
  • $\begingroup$ Step 18 of Machinelearningplus answers your question, I think $\endgroup$ – Emil Jun 21 '19 at 12:10

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.