Using Partial Seeded Topic Modelling

[ghost]

Hello, I'm trying to run the seeded topic modeling for 40 topics but just seeding 20 topics and allowing the algorithm to learn other 20 topics. However, I'm unable to figure out which topics are seeded topics and which are not seeded topics in the end output. It would be very helpful if you have a simple work around in the package. Thanks

[vi3k6i5]

Hey Dipin,

I think when we seed topics we provide the topic id to the words/terms. This should help us figure out the seeded topics and the non seeded ones.

FYI: topic ids starts from 0.

Regards, Vikash

On Fri, Jun 15, 2018 at 2:44 AM goyaldipin1989 [email protected] wrote:

Hello, I'm trying to run the seeded topic modeling for 40 topics but just seeding 20 topics and allowing the algorithm to learn other 20 topics. However, I'm unable to figure out which topics are seeded topics and which are not seeded topics in the end output. It would be very helpful if you have a simple work around in the package. Thanks

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/vi3k6i5/GuidedLDA/issues/17, or mute the thread https://github.com/notifications/unsubscribe-auth/AC-NwhJHFdg7jzTOJCOMDM_y4ASuEfAJks5t8tIngaJpZM4UorBY .