Automatic Interpretation of the English Possessive
Tratz, Stephen and Hovy, Eduard

Article Structure

Abstract

The English ’5 possessive construction occurs frequently in text and can encode several different semantic relations; however, it has received limited attention from the computational linguistics community.

Introduction

The English ’5 possessive construction occurs frequently in text—approximately 1.8 times for every 100 hundred words in the Penn Treebank1 (Marcus et al., l993)—and can encode a number of different semantic relations including ownership (John ’5 car), part-of-whole (John ’5 arm), extent (6 hours’ drive), and location (America ’s rivers).

Background

Although the linguistics field has devoted significant effort to the English possessive (§6.l), the computational linguistics community has given it limited attention.

Dataset Creation

We created the dataset used in this work from three different sources, each representing a distinct genre—newswire, nonfiction, and fiction.

Semantic Relation Inventory

The initial semantic relation inventory for possessives was created by first examining some of the relevant literature on possessives, including work by Badulescu and Moldovan (2009), Barker (1995), Quirk et al.

Experiments

For the automatic classification experiments, we set aside 10% of the data for test purposes, and used the the remaining 90% for training.

Related Work

6.1 Linguistics

Conclusion

In this paper, we present a semantic relation inventory for ’s possessives consisting of 17 relations expressed by the English ’5 construction, the

Future Work

Going forward, we would like to examine the various ambiguities of possessives described in Section 4.3.

Topics

semantic relations

Appears in 19 sentences as: Semantic relation (1) semantic relation (6) semantic relations (12)
In Automatic Interpretation of the English Possessive
  1. The English ’5 possessive construction occurs frequently in text and can encode several different semantic relations ; however, it has received limited attention from the computational linguistics community.
    Page 1, “Abstract”
  2. This paper describes the creation of a semantic relation inventory covering the use of ’s, an inter-annotator agreement study to calculate how well humans can agree on the relations, a large collection of possessives annotated according to the relations, and an accurate automatic annotation system for labeling new examples.
    Page 1, “Abstract”
  3. The English ’5 possessive construction occurs frequently in text—approximately 1.8 times for every 100 hundred words in the Penn Treebank1 (Marcus et al., l993)—and can encode a number of different semantic relations including ownership (John ’5 car), part-of-whole (John ’5 arm), extent (6 hours’ drive), and location (America ’s rivers).
    Page 1, “Introduction”
  4. These interpretations could be valuable for machine translation to or from languages that allow different semantic relations to be encoded by
    Page 1, “Introduction”
  5. This paper presents an inventory of 17 semantic relations expressed by the English ’s—construction, a large dataset annotated according to the this inventory, and an accurate automatic classification system.
    Page 1, “Introduction”
  6. Badulescu and Moldovan (2009) investigate both ’s-constructions and 0f constructions in the same context using a list of 36 semantic relations (including OTHER).
    Page 1, “Background”
  7. For the 960 extracted ’s—possessive examples, only 20 of their semantic relations are observed, including OTHER, with 8 of the observed relations occurring fewer than 10 times.
    Page 1, “Background”
  8. Also, it is sometimes difficult to understand the meaning of the semantic relations , partly because most relations are only described by a single example and, to a lesser extent, because the bulk of the given examples are of-constructions.
    Page 2, “Background”
  9. Table l: The 20 (out of an original 36) semantic relations observed by Badulescu and Moldovan (2009) along with their examples.
    Page 2, “Background”
  10. The initial semantic relation inventory for possessives was created by first examining some of the relevant literature on possessives, including work by Badulescu and Moldovan (2009), Barker (1995), Quirk et al.
    Page 2, “Semantic Relation Inventory”
  11. Table 2: The semantic relations proposed by Quirk et al.
    Page 2, “Semantic Relation Inventory”

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

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

Back to top.

WordNet

Appears in 8 sentences as: WordNet (8)
In Automatic Interpretation of the English Possessive
  1. Many of these templates utilize information from WordNet (Fell-baum, 1998).
    Page 5, “Experiments”
  2. 0 WordNet link types (link type list) (e.g., attribute, hypernym, entailment)
    Page 5, “Experiments”
  3. 0 Lexicographer filenames (lexnames)—top level categories used in WordNet (e.g., noun.body, verb.cognition)
    Page 5, “Experiments”
  4. 0 Set of words from the WordNet definitions (gloss terms)
    Page 5, “Experiments”
  5. 0 The list of words connected via WordNet part-of links (part words)
    Page 5, “Experiments”
  6. 0 List of all possible parts-of-speech in WordNet for the word
    Page 5, “Experiments”
  7. ° WordNet hypernyms
    Page 5, “Experiments”
  8. ° WordNet synonyms
    Page 5, “Experiments”

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

See all papers in Proc. ACL that mention WordNet.

Back to top.

Penn Treebank

Appears in 7 sentences as: Penn Treebank (7)
In Automatic Interpretation of the English Possessive
  1. 21,938 total examples, 15,330 come from sections 2—21 of the Penn Treebank (Marcus et al., 1993).
    Page 2, “Dataset Creation”
  2. For the Penn Treebank , we extracted the examples using the provided gold standard parse trees, whereas, for the latter cases, we used the output of an open source parser (Tratz and Hovy, 2011).
    Page 2, “Dataset Creation”
  3. Penn Treebank , respectively.
    Page 3, “Semantic Relation Inventory”
  4. portion of the Penn Treebank .
    Page 4, “Semantic Relation Inventory”
  5. The Penn Treebank and The History of the Decline and Fall of the R0-man Empire were substantially more similar, although there are notable differences.
    Page 4, “Semantic Relation Inventory”
  6. The accuracy figures for the test instances from the Penn Treebank , The Jungle Book, and The History of the Decline and Fall of the Roman Empire were 88.8%, 84.7%, and 80.6%, respectively.
    Page 5, “Experiments”
  7. The NomBank project (Meyers et al., 2004) provides coarse annotations for some of the possessive constructions in the Penn Treebank , but only those that meet their criteria.
    Page 8, “Related Work”

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

Appears in 7 sentences as: Treebank (7)
In Automatic Interpretation of the English Possessive
  1. 21,938 total examples, 15,330 come from sections 2—21 of the Penn Treebank (Marcus et al., 1993).
    Page 2, “Dataset Creation”
  2. For the Penn Treebank , we extracted the examples using the provided gold standard parse trees, whereas, for the latter cases, we used the output of an open source parser (Tratz and Hovy, 2011).
    Page 2, “Dataset Creation”
  3. Penn Treebank , respectively.
    Page 3, “Semantic Relation Inventory”
  4. portion of the Penn Treebank .
    Page 4, “Semantic Relation Inventory”
  5. The Penn Treebank and The History of the Decline and Fall of the R0-man Empire were substantially more similar, although there are notable differences.
    Page 4, “Semantic Relation Inventory”
  6. The accuracy figures for the test instances from the Penn Treebank , The Jungle Book, and The History of the Decline and Fall of the Roman Empire were 88.8%, 84.7%, and 80.6%, respectively.
    Page 5, “Experiments”
  7. The NomBank project (Meyers et al., 2004) provides coarse annotations for some of the possessive constructions in the Penn Treebank , but only those that meet their criteria.
    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.

feature templates

Appears in 3 sentences as: feature template (1) feature templates (2)
In Automatic Interpretation of the English Possessive
  1. We used 5-fold cross-validation performed using the training data to tweak the included feature templates and optimize training parameters.
    Page 5, “Experiments”
  2. The following feature templates are used to generate features from the above words.
    Page 5, “Experiments”
  3. To evaluate the importance of the different types of features, the same experiment was rerun multiple times, each time including or excluding exactly one feature template .
    Page 6, “Experiments”

See all papers in Proc. ACL 2013 that mention feature templates.

See all papers in Proc. ACL that mention feature templates.

Back to top.

hypernyms

Appears in 3 sentences as: hypernym (1) hypernyms (2)
In Automatic Interpretation of the English Possessive
  1. 0 WordNet link types (link type list) (e.g., attribute, hypernym , entailment)
    Page 5, “Experiments”
  2. ° WordNet hypernyms
    Page 5, “Experiments”
  3. Curiously, although hypernyms are commonly used as features in NLP classification tasks, gloss terms, which are rarely used for these tasks, are approximately as useful, at least in this particular context.
    Page 7, “Experiments”

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

See all papers in Proc. ACL that mention hypernyms.

Back to top.

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