Keynotes

Shallow semantic analyzers, such as semantic role labeling and sense tagging, are increasing in accuracy and becoming commonplace. However, they only provide limited and local representations of words and individual predicate-argument structures. This talk will address some of the current challenges in producing deeper, connected representations of eventualities. Available resources, such as VerbNet, FrameNet, and TimeBank, that can assist in this process, will also be discussed, as well as some of their limitations.