Open mcelvg opened 7 years ago
Is Semantic Role Labeling so difficult task to solve with appropriate quality? Just curious why it took so much time to release?
Sorry for the late reply; it's not the matter of difficulty as SRL used to be a part of ClearNLP, the predecessor of NLP4J. There are two things going on with NLP4J at the moment, transitions to deep learning and graph-based structure parsing. The 2nd part is the reason for delaying SRL; I'm preparing a parser that generates deep dependency structures which could potentially replace SRL. On the side note, I think the field is moving towards to more semantic parsing or AMR parsing instead of SRL, and that's another reason of not investigating much time on SRL. Sorry for the delay.
Good to hear that.
Are there still plans to support semantic role labeling? New date for release? https://emorynlp.github.io/nlp4j/release.html
Any tasks others could help with?