Something went wrong!
Loading...
poster
Building Large-Scale English and Korean Datasets for Aspect-Level Sentiment Analysis in Automotive Domain
Speakers

Dongmin Hyun
Post-doc @ POSTECH
Please log in to leave a comment

Maria Sukhareva
about 4 years agoAnd a follow up question, did you use multilingual BERT for both Korean and English? Why do you think BERT does not perform as well on Korean?
0

Maria Sukhareva
about 4 years agoHi, Thank you for your work! Do you have an explanation why interannotator agreement of the crowdsourced English data is much lower than of Korean?
0