PropBank

The Original PropBank

The original PropBank project, funded by ACE, created a corpus of text annotated with information about basic semantic propositions. Predicate-argument relations were added to the syntactic trees of the Penn Treebank. This resource is now available via LDC.

PropBank Today

This project was continued under NSF funding and DARPA GALE and BOLT. with the aim of creating Parallel PropBanks (the English-Chinese Treebank/PropBank, as well as Arabic) and also PropBanking other genres, such as Broadcast News, Broadcast Conversation, WebText and Discussion Fora, at the University of Colorado. Domain specific annotation of Biomedical data and clinical data has been added. PropBank is also mapped to VerbNet and FrameNet as part of SemLink: Mapping together PropBank/VerbNet/FrameNet. PropBank's coverage has been extended to provide support for AMR and UMR annotation, which make heavy use of PropBank frame files. This was also funded by NSF and DARPA DEFT and BOLT. For up-to-date PropBank data and guidelines, visit the PropBank website.