This dataset contains question-answer pairs to model the predicate-argument structure of deverbal nominalizations. The questions start with wh-words (Who, What, Where, What, etc.) and contain the verbal form of a nominalization from the sentence; the answers are phrases in the sentence.
See the paper for details: QANom: Question-Answer driven SRL for Nominalizations (Klein et. al., COLING 2020)
For previewing the QANom data along with the verbal annotations of QASRL, check out https://browse.qasrl.org/. Also check out our GitHub repository to find code for nominalization identification, QANom annotation, evaluation, and models.
The dataset was annotated by selected workers from Amazon Mechanical Turk.
This dataset contains question-answer pairs to model the predicate-argument structure of deverbal nominalizations. The questions start with wh-words (Who, What, Where, What, etc.) and contain the verbal form of a nominalization from the sentence; the answers are phrases in the sentence.
See the paper for details: QANom: Question-Answer driven SRL for Nominalizations (Klein et. al., COLING 2020)
For previewing the QANom data along with the verbal annotations of QASRL, check out https://browse.qasrl.org/. Also check out our GitHub repository to find code for nominalization identification, QANom annotation, evaluation, and models.
The dataset was annotated by selected workers from Amazon Mechanical Turk.
