This reminds me of this recent working paper, where they try to define what optimal news is (in order to be able to measure if it is biased). I suspect that outside of a few, well-defined, use cases, your measurement will be hopelessly vague. It is still an important goal.
This reminds me of this recent working paper, where they try to define what optimal news is (in order to be able to measure if it is biased). I suspect that outside of a few, well-defined, use cases, your measurement will be hopelessly vague. It is still an important goal.
https://www.nber.org/papers/w32512
Interesting paper, it also references https://pubs.aeaweb.org/doi/pdfplus/10.1257/mic.20190133, which seems even more similar.