‘Writing good templates for automated journalism is not a trivial task, and it can be uncomfortable for journalists to see their craft deconstructed into algorithmically-assembled blocks.’ 

BBC News Labs calls its data-driven news system as ‘semi-automated journalism.’ It keeps human journalists in the loop for approvals before publication.

The objective is to increase local story output without ‘substantial’ increase in workload.

The project is named SALCO, for ‘semi-automated local content.’ Their workflow has five main pieces:

  1. process data and extract bits of interest
  2. produce a text story based on a template prepared by a journalist
  3. generate a graphic that summarises the data in house style
  4. preview for approval by a journalist
  5. publish to appropriate pages

The story goes into detail for each of these stages.

Among the key things learned in the project is the importance of leaving the journalists to write the templates.

SEE FULL STORY

Stories by numbers: How BBC News is experimenting with semi-automated journalism
BBC NEWS LABS | March 22, 2019 | by Roo Hutton

LATEST