Natural Language Understanding Wiki
Register
Advertisement

This article collects notes on reading and reproducing (part of) Silberer and Frank (2012)[1].

Footnote 2: "The phrase type must be NPB, S, VP, SBAR, or SG." According to this list:

  • S: "simple declarative clause, i.e. one that is not introduced by a (possibly empty) subordinating conjunction or a wh-word and that does not exhibit subject-verb inversion."
  • VP: "verb Phrase"
  • SBAR: "Clause introduced by a (possibly empty) subordinating conjunction"

NPB is a tag that does not exist in OntoNotes (proof). I found it as a preprocessing step in another paper (Vadas and Curran, 2011)[2]:

"a preprocessing step is taken wherein NP brackets that do not dominate any other non-possessive NP nodes are relabeled as NPB. For consistency, an extra NP bracket is inserted around NPB nodes not already dominated by an NP. These NPB nodes are removed before evaluation. An example of this transformation can be seen here:"

(S(NP (DT The) (NN dog) )

(VP (VBZ barks) ) )

(S (NP

(NPB (DT The) (NN dog) ) )

(VP (VBZ barks) ) )

Collins-2003-sg

Figure from Collins (2003) demonstrating the creation of SG nodes.

SG also doesn't exist in OntoNotes (proof) either. Collins (2003)[3] is a candidate:

"... sentences with and without subjects appear in quite different syntactic environments. For these reasons we modify the nonterminal for sentences without subjects to be SG"

Evaluation metric: "We adopt the precision (P), recall (R) and F1 measures in Ruppenhofer et al. (2010)."

References[]

  1. Silberer, C., & Frank, A. (2012). Casting Implicit Role Linking As an Anaphora Resolution Task. In Proceedings of the First Joint Conference on Lexical and Computational Semantics - Volume 1: Proceedings of the Main Conference and the Shared Task, and Volume 2: Proceedings of the Sixth International Workshop on Semantic Evaluation (pp. 1–10). Stroudsburg, PA, USA: Association for Computational Linguistics.
  2. Vadas, D., & Curran, J. R. (2011). Parsing Noun Phrases in the Penn Treebank. Computational Linguistics, 37(4), 753–809. http://doi.org/10.1162/COLI_a_00076
  3. Collins, M. (2003). Head-Driven Statistical Models for. Journal of Computational Linguistics, 29(4), 589–637. http://doi.org/10.1162/089120103322753356
Advertisement