I am new to text analysis and am stuck on a question that doesn't seem to be answered in the documentation (or at least, I can't find it).
I have created an STM in R from a Quanteda DfM which has docvars associated to it. The topics are based on news sources and the docvars indicate the news outlet and date (as one string).
The STM estimation is going fine, but for my project what I really want to know is which news outlet and date combo is associated to which topics. Essentially I want an output like:
outlet_date topic_1 topic_2 topic_3
newyorktimes_20181001 0.6 0.23 0.17
newyorktimes_20181002 0.54 0.4 0.06
newyorktimes_20181003 0.2 0.73 0.07
Is this possible?
Sorry for not having a reproducible example - not sure how to include one given all the steps to get to this point. Thanks so much!