Selectional Preferences for Semantic Role Classification
Bibliographic Data
| ID | 12155802 |
|---|---|
| Authors | Beñat Zapirain (University of the Basque Country, corresponding author), Eneko Agirre (0000-0002-0195-4899, University of the Basque Country), Lluı́s Màrquez (Universitat Politècnica de Catalunya), Lluís Màrquez (Universitat Politècnica de Catalunya), Mihai Surdeanu (0000-0001-6956-8030, University of Arizona) |
| Year | 2012 |
| Volume | 39 |
| Issue | 3 |
| Pages | 631-663 |
| Publication date | 2012-11-16 |
| Peer Reviewed | Yes |
| Open Access | Yes |
| Type | ARTICLE |
| Venue | Computational Linguistics (JOURNAL) |
| Journal identifiers | ISSN: 0891-2017 • E-ISSN: 1530-9312 |
| Publisher | Association for Computational Linguistics (PUBLISHER • US) |
| DOI | 10.1162/coli_a_00145 |
| OpenAlex | W2080635174 |
| Language | EN |
| Citations received | 4 |
| References cited | 34 |
This paper focuses on a well-known open issue in Semantic Role Classification (SRC) research: the limited influence and sparseness of lexical features. We mitigate this problem using models that integrate automatically learned selectional preferences (SP). We explore a range of models based on WordNet and distributional-similarity SPs. Furthermore, we demonstrate that the SRC task is better modeled by SP models centered on both verbs and prepositions, rather than verbs alone. Our experiments with SP-based models in isolation indicate that they outperform a lexical baseline with 20 F 1 points in domain and almost 40 F 1 points out of domain. Furthermore, we show that a state-of-the-art SRC system extended with features based on selectional preferences performs significantly better, both in domain (17% error reduction) and out of domain (13% error reduction). Finally, we show that in an end-to-end semantic role labeling system we obtain small but statistically significant improvements, even though our modified SRC model affects only approximately 4% of the argument candidates. Our post hoc error analysis indicates that the SP-based features help mostly in situations where syntactic information is either incorrect or insufficient to disambiguate the correct role
Argument (complex analysis · Baseline (sea · Domain (mathematical analysis · Domain adaptation · Machine learning · Natural language processing · Reduction (mathematics · Semantic similarity · Task (project management · Word error rate · WordNet · Computer Science · Mathematics · Natural Language Processing Techniques · Text Readability and Simplification · Topic Modeling · Artificial Intelligence
Class-Based Probability Estimation Using a Semantic Hierarchy
Semantic Role Labeling
Learning to Rank Answers to Non-Factoid Questions from Web Collections
Disambiguating Nouns, Verbs, and Adjectives Using Automatically Acquired Selectional Preferences
Towards Robust Semantic Role Labeling
A Flexible, Corpus-Driven Model of Regular and Inverse Selectional Preferences
Dependency-Based Construction of Semantic Space Models
The Proposition Bank
Automatic Labeling of Semantic Roles
Distributional Memory
| Unique citing works | 4 |
|---|---|
| Citations per year | 0,22 |
| Citation span | 2008 - 2011 (4) |
| Citation velocity | historical |
| Highly cited | No |
| Citation types | Neutral: 4 |