Skip to content

Latest commit

 

History

History
18 lines (14 loc) · 1.66 KB

readme.md

File metadata and controls

18 lines (14 loc) · 1.66 KB

Contents

Code and instructions for applying BERTopic Topic Modelling to free text survey responses from a survey of people's lived experiences during extreme weather events in Bristol. No data is stored in this repo.

BertTopic.ipynb

  • This notebook loads in your data, trains your BERTopic model and saves some outputs including data tables and diagnostic plots.
  • The data loading is left to give an example but will not run unless you have the required datasets in your working directory.
  • They can however be easily modified to load in other datasets.
  • This BERTopic model has commented out code that means you can run this as a ZeroShot model (you can dictate zeroshot terms in the data loading stage or manually deploy them) but it also runs fine unsupervised.

icicleplots.ipynb

  • This notebook loads in documents and their topics, puts them into custom categories (as defined by Eunice and Jo).
  • Then outputs icicle plots with some customisation using plotly.

TopicModellingWithBERT.pptx

  • Slide deck for a talk Huw gave to the Bristol Climate group on 13/11/2024.
  • Tried to keep text sparse on the slides but more denser in the notes so you can follow along!
  • Recording of the Zoom talk can be found here