"This repo contains code for the paper, 'Automatically Neutralizing Subjective Bias in Text'.
Concretely this means algorithms for;
- Identifying biased words in sentences.
- Neutralizing bias in sentences.
E.g.
'John McCain exposed as an unprincipled politican' -> 'John McCain described as an unprincipled politican'
"
This idea could have interesting applications. Imagine it being a UI option in twitter or the browser to 'unopinionate' the text, leaving users more room to form their own conclusions. This would probably reduce the emotionally driven engagement metrics of these platforms, but could it help having a healthier collective discourse?
This is great! If you're interested in this kind of bias correction for writing, there's a volunteer team named "Write for All" that is seeking help here: http://writeforall.org (feel free to message me about it).
3 comments
[ 3.1 ms ] story [ 19.9 ms ] thread"This repo contains code for the paper, 'Automatically Neutralizing Subjective Bias in Text'.
Concretely this means algorithms for; - Identifying biased words in sentences. - Neutralizing bias in sentences.
E.g. 'John McCain exposed as an unprincipled politican' -> 'John McCain described as an unprincipled politican' "
This idea could have interesting applications. Imagine it being a UI option in twitter or the browser to 'unopinionate' the text, leaving users more room to form their own conclusions. This would probably reduce the emotionally driven engagement metrics of these platforms, but could it help having a healthier collective discourse?