Universal Conceptual Cognitive Annotation (UCCA)
Abend, Omri and Rappoport, Ari

Article Structure

Abstract

Syntactic structures, by their nature, reflect first and foremost the formal constructions used for expressing meanings.

Introduction

Syntactic structures are mainly committed to representing the formal patterns of a language, and only indirectly reflect semantic distinctions.

The UCCA Scheme

2.1 The Formalism

A UCCA-Annotated Corpus

The annotated text is mostly based on English Wikipedia articles for celebrities.

UCCA’s Benefits to Semantic Tasks

UCCA’s relative insensitivity to syntactic forms has potential benefits for a wide variety of seman-

Related Work

In this section we compare UCCA to some of the major approaches to grammatical representation in NLP.

Conclusion

This paper presented Universal Conceptual Cognitive Annotation (UCCA), a novel framework for semantic representation.

Topics

semantic roles

Appears in 5 sentences as: Semantic role (1) semantic role (1) semantic roles (3)
In Universal Conceptual Cognitive Annotation (UCCA)
  1. (2012) for semantic role labeling and Kwiatkowski et al.
    Page 8, “Related Work”
  2. Semantic role labeling (SRL) schemes bear similarity to the foundational layer, due to their focus on argument structure.
    Page 8, “Related Work”
  3. PropBank and NomBank are built on top of the PTB annotation, and provide for each verb (PropBank) and noun (NomBank), a delineation of their arguments and their categorization into semantic roles .
    Page 9, “Related Work”
  4. proposes a comprehensive approach to semantic roles .
    Page 9, “Related Work”
  5. It defines a lexical database of Frames, each containing a set of possible frame elements and their semantic roles .
    Page 9, “Related Work”

See all papers in Proc. ACL 2013 that mention semantic roles.

See all papers in Proc. ACL that mention semantic roles.

Back to top.

coreference

Appears in 4 sentences as: coreference (5)
In Universal Conceptual Cognitive Annotation (UCCA)
  1. Unlike common practice in grammatical annotation, linkage relations in UCCA can cross sentence boundaries, as can relations represented in other layers (e.g., coreference ).
    Page 4, “The UCCA Scheme”
  2. Another immediate extension to UCCA’s foundational layer can be the annotation of coreference relations.
    Page 5, “The UCCA Scheme”
  3. A coreference layer would annotate a relation between “John” and “his” by introducing a new node whose descendants are these two units.
    Page 5, “The UCCA Scheme”
  4. The fact that this node represents a coreference relation would be represented by a label on the edge connecting them to the coreference node.
    Page 5, “The UCCA Scheme”

See all papers in Proc. ACL 2013 that mention coreference.

See all papers in Proc. ACL that mention coreference.

Back to top.

F-score

Appears in 4 sentences as: F-score (5)
In Universal Conceptual Cognitive Annotation (UCCA)
  1. We derive an F-score from these counts.
    Page 6, “A UCCA-Annotated Corpus”
  2. The table presents the average F-score between the annotators, as well as the average F-score when comparing to the gold standard.
    Page 6, “A UCCA-Annotated Corpus”
  3. An average taken over a sample of passages annotated by all four annotators yielded an F-score of 93.7%.
    Page 6, “A UCCA-Annotated Corpus”
  4. A recent work that did report inter-annotator agreement in terms of bracketing F-score is an annotation project of the PTB’s noun phrases with more elaborate syntactic structure (Vadas and Cur-
    Page 6, “A UCCA-Annotated Corpus”

See all papers in Proc. ACL 2013 that mention F-score.

See all papers in Proc. ACL that mention F-score.

Back to top.

gold standard

Appears in 4 sentences as: gold standard (3) “gold standard” (1)
In Universal Conceptual Cognitive Annotation (UCCA)
  1. In addition, a “gold standard” was annotated by the authors of this paper.
    Page 6, “A UCCA-Annotated Corpus”
  2. The table presents the average F-score between the annotators, as well as the average F-score when comparing to the gold standard .
    Page 6, “A UCCA-Annotated Corpus”
  3. The obtained F-scores when comparing to a gold standard , ordered decreasingly according to the annotator’s acquaintance with linguistics, were 78%, 74.4%, 69.5% and 67.8%.
    Page 6, “A UCCA-Annotated Corpus”
  4. Indeed, the obtained F-scores, again compared to a gold standard and averaged over the next five training passages, were (by the same order) 78.6%, 77.3%, 79.2% and 78%.
    Page 6, “A UCCA-Annotated Corpus”

See all papers in Proc. ACL 2013 that mention gold standard.

See all papers in Proc. ACL that mention gold standard.

Back to top.

machine translation

Appears in 4 sentences as: machine translation (4)
In Universal Conceptual Cognitive Annotation (UCCA)
  1. One example is machine translation to target languages that do not express this structural distinction (e.g., both (a) and (b) would be translated to the same German sentence “John duschte”).
    Page 1, “Introduction”
  2. Aside from machine translation , a great variety of semantic tasks can benefit from a scheme that is relatively insensitive to syntactic variation.
    Page 8, “UCCA’s Benefits to Semantic Tasks”
  3. A different strand of work addresses the construction of an interlingual representation, often with a motivation of applying it to machine translation .
    Page 8, “Related Work”
  4. We are currently attempting to construct a parser for UCCA and to apply it to several semantic tasks, notably English-French machine translation .
    Page 9, “Conclusion”

See all papers in Proc. ACL 2013 that mention machine translation.

See all papers in Proc. ACL that mention machine translation.

Back to top.

semantic representation

Appears in 4 sentences as: semantic representation (4)
In Universal Conceptual Cognitive Annotation (UCCA)
  1. We present UCCA, a novel multilayered framework for semantic representation that aims to accommodate the semantic distinctions expressed through linguistic utterances.
    Page 1, “Abstract”
  2. An extensive comparison of UCCA to existing approaches to syntactic and semantic representation , focusing on the major resources available for English, is found in Section 5.
    Page 2, “Introduction”
  3. Several annotated corpora offer a joint syntactic and semantic representation .
    Page 8, “Related Work”
  4. This paper presented Universal Conceptual Cognitive Annotation (UCCA), a novel framework for semantic representation .
    Page 9, “Conclusion”

See all papers in Proc. ACL 2013 that mention semantic representation.

See all papers in Proc. ACL that mention semantic representation.

Back to top.

Treebank

Appears in 4 sentences as: Treebank (4) treebank (1)
In Universal Conceptual Cognitive Annotation (UCCA)
  1. In fact, the annotations of (a) and (c) are identical under the most widely-used schemes for English, the Penn Treebank (PTB) (Marcus et al., 1993) and CoNLL-style dependencies (Surdeanu et al., 2008) (see Figure l).
    Page 1, “Introduction”
  2. For instance, both the PTB and the Prague Dependency Treebank (Bo'hmova et al., 2003) employed annotators with extensive linguistic background.
    Page 6, “A UCCA-Annotated Corpus”
  3. The most prominent annotation scheme in NLP for English syntax is the Penn Treebank .
    Page 8, “Related Work”
  4. Examples include the Groningen Meaning bank (Basile et al., 2012), Treebank Semantics (Butler and Yoshi-moto, 2012) and the Lingo Redwoods treebank (Oepen et al., 2004).
    Page 8, “Related Work”

See all papers in Proc. ACL 2013 that mention Treebank.

See all papers in Proc. ACL that mention Treebank.

Back to top.

PDF available from www.aclweb.org.
Send us an email