A package? boilerplate code? for Izzan's topic modelling pipeline using K-means clustering, BERT-based transformers, and Indonesian-specific text preprocessing. Enjoy!
Inspired: OCTIS: Comparing and Optimizing Topic models is Simple! (Terragni et al., EACL 2021)