103 - Processing Language in Social Media, with Brendan O'Connor
NLP Highlights
00:00
Is There a Bias in Social Media?
The big bias that everyone points to immediately is just who the heck are the authors that you're analyzing, especially when you're dealing with a system of twitters like this. Arguably, this is why malicious actors can push misinformation. And i think we don't know at a technical level what's a good way of combining multi-lingual sentiment on social media. What does it mean to do that fairly? Is a really interesting question,. and there's many different ways ton code fairness there.
Transcript
Play full episode
