Emerging Thoughts on Benchmarking
The following would be natural places to start. For most of these one would need to find suitable code for existing BERT-based architectures for e.g. English. For the first though, document-level SA on NoReC, Jeremy would have an existing set-up for using mBERT.
- NoReC; for document-level sentiment analysis (i.e. rating prediction).
- NoReC_fine; for fine-grained sentiment analysis (e.g. predicting target expression + polarity)
- NDT; for dependency parsing or PoS tagging (perhaps best to use the UD version)
- NorNE; for named entity recognition, extends NDT (also available for the UD version)