Statistical optimization of a topic model

This workflow shows how to use and interpret a test for the multinomial assumption overlying LDA model. The selected structure is the one that most closely matches the word proportions in the underlying corpus Useful to choose correctly the number of topics that characterize the corpus. First part of the worklow done by adewi in 17_TopicExtraction_with_the_ElbowMethod.


This is a companion discussion topic for the original entry at https://kni.me/w/EPXd3E9GhMaZUns7