I am making SMS categorizer. For this I want to classify my messages into different topics. So I want to use gensim for that. Can anybody provide me the source of any tutorial that can help me to begin topic modelling using gensim?
1 Answers
Radim Řehůřek, the author of gensim, provides tutorials on his website, which I found quite helpful: https://radimrehurek.com/gensim/tutorial.html
He also provides some tutorials on his github that are not that easy to find, so digging for a while in there might bring some great notebooks to the surface. For example, these two (for LDA): https://github.com/RaRe-Technologies/gensim/blob/develop/docs/notebooks/lda_training_tips.ipynb
https://github.com/RaRe-Technologies/gensim/blob/develop/docs/notebooks/topic_methods.ipynb
I personally found it very thrilling to start with the English Wikipedia, for which Řehůřek also provides a tutorial. Even though it takes quite a long time to train the model, you will find yourself with very "natural" seeming topics, which make it easy to test gensim's functionality.