3 ms·
Good points! I'll keep that in mind in what I hope is already a relatively cautious approach to that sort of comment. On the other hand, ditching all/most idiom
by ineedasername 6y ago
Good points! I'll keep that in mind in what I hope is already a relatively cautious approach to that sort of comment. On the other hand, ditching all/most idiomatic English and subtleties is probably not the right approach either, or avoiding all cultural references because others might not understand-- I don't think you're suggesting that, it's just my observation that this issue exists on a spectrum and not a binary.
Edit: You're in AI-- what's the landscape look like for improving automated translation to detect sarcasm? As a problem domain, it seems like it would be a subset/cross-section of sentiment analysis & word-sense-disambiguation.
- nl 6y agoThese days the biggest issue is getting a decent labelled dataset. Once you have that you just fine-tune a BERT-dervived transformer neural network. See https://paperswithcode.com/task/sarcasm-detection/codeless https://paperswithcode.com/task/sarcasm-detection/codeless for some examples and datasets.