OxfordDemSci / ICS_Analysis

Mixed methods approach and interactive dashboard to analyse research impact through Impact Case Studies submitted to the UK's Research Excellence Framework (REF) 2021.
https://shape-impact.co.uk
GNU General Public License v3.0
5 stars 0 forks source link

Move text_helpers up in the pipeline. #4

Closed crahal closed 1 year ago

crahal commented 1 year ago

We need to move the text_helpers.py stuff (e.g. tokenizing/lemmatizing further up the pipeline, ideally into the area of code that gets feature engineered. We can create five seperate fields (similar to s1*-s5* as BZ has done with Sentiment and Readability scores.

crahal commented 1 year ago

Closing this as deciding not to do it; topic modelling has become entirely seperated from the (co-)occurance analysis.