Article Structure
Abstract
Recent work on statistical quantifier scope disambiguation (QSD) has improved upon earlier work by scoping an arbitrary number and type of noun phrases.
Introduction
The sentence there is one faculty member in every graduate committee is ambiguous with respect to quantifier scoping, since there are at least two possible readings: If one has wide scope, there is a unique faculty member on every committee.
Task definition
In QuanText, scope-bearing elements (or, as we call them, scopal terms) of each sentence have been identified using labeled chunks, as in (3).
Our framework
3.1 Learning to do QSD
Experiments
QuanText contains 500 sentences with a total of 1750 chunks, that is 3.5 chunks/sentence on average.
Related work
Since automatic QSD is in general challenging, traditionally quantifier scoping is left underspecified in deep linguistic processing systems (Al-shawi and Crouch, 1992; Bos, 1996; Copestake et al., 2001).
Summary and future work
We develop the first statistical QSD model addressing the interaction of quantifiers with negation and the implicit universal of plurals, defining a baseline for this task on QuanText data (Manshadi et al., 2012).
Topics
dependency relations
- Although regular “untyped” dependency relations do not seem to help our QSD system in the presence of phrase-structure trees, we found the col-Page 6, “Our framework”
- 0 Type of incoming dependency relation of each noun 0 Syntactic category of the deepest common ancestor o Lexical item of the deepest common ancestor 0 Length of the undirected path between the twoPage 6, “Our framework”
- 14MAll’s features are similar to part-of-speech tags and untyped dependency relations .Page 7, “Experiments”
- dependency relations ) are used on both the train and the test data.Page 8, “Related work”
See all papers in Proc. ACL 2013 that mention dependency relations.
See all papers in Proc. ACL that mention dependency relations.
Back to top.
baseline system
- However, in order to have a fair comparison, we have used the output of the Stanford parser to automatically generate the same features that MAll have hand-annotated.14 In order to run the baseline system on implicit universals, we take the feature vector of a plural NP and add a feature to indicate that this feature vector represents the implicit universal of the corresponding chunk.Page 7, “Experiments”
- Once again, in order to have a fair comparison, we apply a similar modification to the baseline system .Page 7, “Experiments”
- We also use the exact same classifier as used in MAl 1.15 Figure 5(a) compares the performance of our model, which we refer to as RPC-SVM-l3, with the baseline system , but only on explicit NP chunks.16 The goal for running this experiment has been to compare the performance of our model to the baseline sys-token, as described by Manshadi et a1.Page 7, “Experiments”
See all papers in Proc. ACL 2013 that mention baseline system.
See all papers in Proc. ACL that mention baseline system.
Back to top.
gold-standard
- Our rich set of features significantly improves the performance of the QSD model, even though we give up the gold-standard dependency features (Sect.Page 2, “Introduction”
- For example if G3 in Figure l is a gold-standard DAG and G1 is a candidate DAG, TC-based metrics count 2 > 3 as another match, even though it is entailed from 2 > 1 and 1 > 3.Page 3, “Task definition”
- 19To find the gain that can be obtained with gold-standard parses, we used MAll’s system with their hand-annotated and the equivalent automatically generated features.Page 8, “Related work”
See all papers in Proc. ACL 2013 that mention gold-standard.
See all papers in Proc. ACL that mention gold-standard.
Back to top.
part-of-speech
- The part-of-speech (POS) tag of the head of chunk The lexical item of the head nounPage 6, “Our framework”
- To extract part-of-speech tags, phrase structure trees, and typed dependencies, we use the Stanford parser (Klein and Manning, 2003; de Marneffe et al., 2006) on both train and test sets.Page 7, “Experiments”
- 14MAll’s features are similar to part-of-speech tags and untyped dependency relations.Page 7, “Experiments”
See all papers in Proc. ACL 2013 that mention part-of-speech.
See all papers in Proc. ACL that mention part-of-speech.
Back to top.
Penn Treebank
- For example, Higgins and Sadock (2003) find fewer than 1000 sentences with two or more explicit quantifiers in the Wall Street journal section of Penn Treebank .Page 1, “Introduction”
- Plurals form 18% of the NPs in our corpus and 20% of the nouns in Penn Treebank .Page 2, “Introduction”
- Explicit universals, on the other hand, form less than 1% of the determiners in Penn Treebank .Page 2, “Introduction”
See all papers in Proc. ACL 2013 that mention Penn Treebank.
See all papers in Proc. ACL that mention Penn Treebank.
Back to top.
Treebank
- For example, Higgins and Sadock (2003) find fewer than 1000 sentences with two or more explicit quantifiers in the Wall Street journal section of Penn Treebank .Page 1, “Introduction”
- Plurals form 18% of the NPs in our corpus and 20% of the nouns in Penn Treebank .Page 2, “Introduction”
- Explicit universals, on the other hand, form less than 1% of the determiners in Penn Treebank .Page 2, “Introduction”
See all papers in Proc. ACL 2013 that mention Treebank.
See all papers in Proc. ACL that mention Treebank.
Back to top.
