Starting from Scratch in Semantic Role Labeling
Connor, Michael and Gertner, Yael and Fisher, Cynthia and Roth, Dan

Article Structure

Abstract

A fundamental step in sentence comprehension involves assigning semantic roles to sentence constituents.

Introduction

In this paper we present experiments with an automatic system for semantic role labeling (SRL) that is designed to model aspects of human language acquisition.

Model

We model language learning as a Semantic Role Labeling (SRL) task (Carreras and Marquez, 2004).

Unsupervised Parsing

As a first step of processing, we feed the learner large amounts of unlabeled text and expect it to learn some structure over this data that will facilitate future processing.

Argument Identification

The unsupervised parser provides a state label for each word in each sentence; the goal of the argument identification stage is to use these states to label words as potential arguments, predicates or neither.

Testing SRL Performance

Finally, we used the results of the previous parsing and argument-identification stages in training a simplified SRL classifier (BabySRL), equipped with sets of features derived from the structure-mapping account.

Conclusion and Future Work

The key innovation in the present work is the combination of unsupervised part-of-speech tagging and argument identification to permit leam-ing in a simplified SRL system.

Topics

semantic roles

Appears in 11 sentences as: Semantic Role (3) semantic role (4) semantic roles (5)
In Starting from Scratch in Semantic Role Labeling
  1. A fundamental step in sentence comprehension involves assigning semantic roles to sentence constituents.
    Page 1, “Abstract”
  2. To accomplish this, the listener must parse the sentence, find constituents that are candidate arguments, and assign semantic roles to those constituents.
    Page 1, “Abstract”
  3. In this paper we focus on the parsing and argument-identification steps that precede Semantic Role Labeling (SRL) training.
    Page 1, “Abstract”
  4. The results show that proposed shallow representations of sentence structure are robust to reductions in parsing accuracy, and that the contribution of alternative representations of sentence structure to successful semantic role labeling varies with the integrity of the parsing and argument-identification stages.
    Page 1, “Abstract”
  5. In this paper we present experiments with an automatic system for semantic role labeling (SRL) that is designed to model aspects of human language acquisition.
    Page 1, “Introduction”
  6. n Semantic Role Labeling
    Page 1, “Introduction”
  7. Previous computational experiments with a system for automatic semantic role labeling (BabySRL: (Connor et al., 2008)) showed that it is possible to learn to assign basic semantic roles based on the shallow sentence representations proposed by the structure-mapping view.
    Page 1, “Introduction”
  8. We model language learning as a Semantic Role Labeling (SRL) task (Carreras and Marquez, 2004).
    Page 2, “Model”
  9. The candidate arguments and predicates identified in each input sentence are passed to an SRL classifier that uses simple abstract features based on the number and order of arguments to learn to assign semantic roles .
    Page 2, “Model”
  10. have the luxury of treating part-of-speech tagging and semantic role labeling as separable tasks.
    Page 9, “Conclusion and Future Work”
  11. An SRL classifier used simple representations built from these identified arguments to extract useful abstract patterns for classifying semantic roles .
    Page 9, “Conclusion and Future Work”

See all papers in Proc. ACL 2010 that mention semantic roles.

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

Back to top.

role labeling

Appears in 9 sentences as: role label (1) Role Labeling (3) role labeling (4) role labels (1)
In Starting from Scratch in Semantic Role Labeling
  1. In this paper we focus on the parsing and argument-identification steps that precede Semantic Role Labeling (SRL) training.
    Page 1, “Abstract”
  2. The results show that proposed shallow representations of sentence structure are robust to reductions in parsing accuracy, and that the contribution of alternative representations of sentence structure to successful semantic role labeling varies with the integrity of the parsing and argument-identification stages.
    Page 1, “Abstract”
  3. In this paper we present experiments with an automatic system for semantic role labeling (SRL) that is designed to model aspects of human language acquisition.
    Page 1, “Introduction”
  4. n Semantic Role Labeling
    Page 1, “Introduction”
  5. Previous computational experiments with a system for automatic semantic role labeling (BabySRL: (Connor et al., 2008)) showed that it is possible to learn to assign basic semantic roles based on the shallow sentence representations proposed by the structure-mapping view.
    Page 1, “Introduction”
  6. We model language learning as a Semantic Role Labeling (SRL) task (Carreras and Marquez, 2004).
    Page 2, “Model”
  7. The stages are: (l) Parsing the sentence, (2) Identifying potential predicates and arguments based on the parse, (3) Classifying role labels for each potential argument relative to a predicate, (4) Applying constraints to find the best labeling of arguments for a sentence.
    Page 2, “Model”
  8. The SRL classifier starts with noisy largely unsupervised argument identification, and receives feedback based on annotation in the PropBank style; in training, each word identified as an argument receives the true role label of the phrase that word is part of.
    Page 2, “Model”
  9. have the luxury of treating part-of-speech tagging and semantic role labeling as separable tasks.
    Page 9, “Conclusion and Future Work”

See all papers in Proc. ACL 2010 that mention role labeling.

See all papers in Proc. ACL that mention role labeling.

Back to top.

semantic role labeling

Appears in 7 sentences as: Semantic Role Labeling (3) semantic role labeling (4)
In Starting from Scratch in Semantic Role Labeling
  1. In this paper we focus on the parsing and argument-identification steps that precede Semantic Role Labeling (SRL) training.
    Page 1, “Abstract”
  2. The results show that proposed shallow representations of sentence structure are robust to reductions in parsing accuracy, and that the contribution of alternative representations of sentence structure to successful semantic role labeling varies with the integrity of the parsing and argument-identification stages.
    Page 1, “Abstract”
  3. In this paper we present experiments with an automatic system for semantic role labeling (SRL) that is designed to model aspects of human language acquisition.
    Page 1, “Introduction”
  4. n Semantic Role Labeling
    Page 1, “Introduction”
  5. Previous computational experiments with a system for automatic semantic role labeling (BabySRL: (Connor et al., 2008)) showed that it is possible to learn to assign basic semantic roles based on the shallow sentence representations proposed by the structure-mapping view.
    Page 1, “Introduction”
  6. We model language learning as a Semantic Role Labeling (SRL) task (Carreras and Marquez, 2004).
    Page 2, “Model”
  7. have the luxury of treating part-of-speech tagging and semantic role labeling as separable tasks.
    Page 9, “Conclusion and Future Work”

See all papers in Proc. ACL 2010 that mention semantic role labeling.

See all papers in Proc. ACL that mention semantic role labeling.

Back to top.

part-of-speech

Appears in 4 sentences as: part-of-speech (4)
In Starting from Scratch in Semantic Role Labeling
  1. The first problem involves classifying words by part-of-speech .
    Page 2, “Introduction”
  2. By using the HMM part-of-speech tagger in this way, we can ask how the simple structural features that we propose children start with stand up to reductions in parsing accuracy.
    Page 2, “Introduction”
  3. The key innovation in the present work is the combination of unsupervised part-of-speech tagging and argument identification to permit leam-ing in a simplified SRL system.
    Page 8, “Conclusion and Future Work”
  4. have the luxury of treating part-of-speech tagging and semantic role labeling as separable tasks.
    Page 9, “Conclusion and Future Work”

See all papers in Proc. ACL 2010 that mention part-of-speech.

See all papers in Proc. ACL that mention part-of-speech.

Back to top.

POS tags

Appears in 4 sentences as: POS tagger (1) POS tagging (1) POS tags (2)
In Starting from Scratch in Semantic Role Labeling
  1. To implement this division into function and content words3, we start with a list of function word POS tags4 and then find words that appear predominantly with these POS tags , using tagged WSJ data (Marcus et al., 1993).
    Page 3, “Unsupervised Parsing”
  2. Smaller numbers are better, indicating less information lost in moving from the HMM states to the gold POS tags .
    Page 4, “Unsupervised Parsing”
  3. We first evaluate these parsers (the first stage of our SRL system) on unsupervised POS tagging .
    Page 4, “Unsupervised Parsing”
  4. When trained on arguments identified via the unsupervised POS tagger , noun pattern features promoted agent interpretations of tran-
    Page 7, “Testing SRL Performance”

See all papers in Proc. ACL 2010 that mention POS tags.

See all papers in Proc. ACL that mention POS tags.

Back to top.

content word

Appears in 3 sentences as: content word (2) content words (1)
In Starting from Scratch in Semantic Role Labeling
  1. During initial unsupervised parsing we experiment with incorporating knowledge through a combination of statistical priors favoring a skewed distribution of words into classes, and an initial hard clustering of the vocabulary into function and content words .
    Page 2, “Model”
  2. Because the function and content word preclus-tering preceded parameter estimation, it can be combined with either EM or VB learning.
    Page 4, “Unsupervised Parsing”
  3. Although this initial split forces sparsity on the emission matrix and allows more uniform sized clusters, Dirichlet priors may still help, if word clusters within the function or content word subsets vary in size and frequency.
    Page 4, “Unsupervised Parsing”

See all papers in Proc. ACL 2010 that mention content word.

See all papers in Proc. ACL that mention content word.

Back to top.

language acquisition

Appears in 3 sentences as: language acquisition (3)
In Starting from Scratch in Semantic Role Labeling
  1. In this paper we present experiments with an automatic system for semantic role labeling (SRL) that is designed to model aspects of human language acquisition .
    Page 1, “Introduction”
  2. Proposed solutions to this problem in the NLP and human language acquisition literatures focus on distributional learning as a key data source (e.g., (Mintz, 2003; Johnson, 2007)).
    Page 2, “Introduction”
  3. The architecture of our system is similar to a previous approach to modeling early language acquisition (Connor et al., 2009), which is itself based on the standard architecture of a full SRL system (e.g.
    Page 2, “Model”

See all papers in Proc. ACL 2010 that mention language acquisition.

See all papers in Proc. ACL that mention language acquisition.

Back to top.

part-of-speech tagging

Appears in 3 sentences as: part-of-speech tagger (1) part-of-speech tagging (2)
In Starting from Scratch in Semantic Role Labeling
  1. By using the HMM part-of-speech tagger in this way, we can ask how the simple structural features that we propose children start with stand up to reductions in parsing accuracy.
    Page 2, “Introduction”
  2. The key innovation in the present work is the combination of unsupervised part-of-speech tagging and argument identification to permit leam-ing in a simplified SRL system.
    Page 8, “Conclusion and Future Work”
  3. have the luxury of treating part-of-speech tagging and semantic role labeling as separable tasks.
    Page 9, “Conclusion and Future Work”

See all papers in Proc. ACL 2010 that mention part-of-speech tagging.

See all papers in Proc. ACL that mention part-of-speech tagging.

Back to top.

word order

Appears in 3 sentences as: word order (3)
In Starting from Scratch in Semantic Role Labeling
  1. Experimental evidence suggests that they do: 21-month-olds mistakenly interpreted word order in sentences such as “The girl and the boy kradded” as conveying agent-patient roles (Gertner and Fisher, 2006).
    Page 1, “Introduction”
  2. We compared the behavior of noun pattern features to another simple representation of word order , position relative to the verb (Veeros).
    Page 6, “Testing SRL Performance”
  3. (2) Because NounPat features represent word order solely in terms of a sequence of nouns, an SRL equipped with these features will make the errors predicted by the structure-mapping account and documented in children (Gertner and Fisher, 2006).
    Page 7, “Testing SRL Performance”

See all papers in Proc. ACL 2010 that mention word order.

See all papers in Proc. ACL that mention word order.

Back to top.

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