Create topical source set about storyweb HOT 6 CLOSED

1. create a GU subset

• filter GU articles on pillar/news & content type article

2. further filter down GU subset

• somehow decide on a keyword, or topic, or label to further narrow down the set DM is analysing the GU data to create a distribution of keywords so we can make a topical selection based on keywords
• run spacy on news subset
• look at textcategorizer pipeline part results
  Default models have no pre-trained textcat component, so this option is off the table.

from storyweb.

@pudo will suggest a keyword from this sheet: https://docs.google.com/spreadsheets/d/1cQPozkD4bpbEcdCdDGfRMiMGfcBNU0dzjAeg4CSQATE/edit#gid=525682012

Overview of GU articles: https://docs.google.com/spreadsheets/d/1LhYL0CTlbOR9KtW8ok15skBqDNesSDMn_3jqtxMCANg/edit#gid=1221445460

from storyweb.

Some guardian keywords I think would make a good starter set:

• law/human-rights
• us-news/us-crime
• business/banking
• technology/cryptocurrencies
• business/shipping-industry
• law/criminal-justice
• business/financial-sector
• money/scamsandfraud
• business/financial-conduct-authority
• world/roman-abramovich
• business/internationaltrade
• business/supply-chain-crisis
• money/tax

from storyweb.

So you suggest making a subset with all of these? Or pick one?

from storyweb.

I mean all of them would be amazing, or even file by file?

from storyweb.

https://docs.google.com/spreadsheets/d/1cQPozkD4bpbEcdCdDGfRMiMGfcBNU0dzjAeg4CSQATE/edit#gid=423272934

from storyweb.