VerbOcean: Mining the Web for Fine-Grained Semantic Verb Relations

By Timothy Chklovski, Patrick Pantel. 2004

507 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Natural Language Processing

Citations in the corpus

A large annotated corpus for learning natural language inference (2015, methodology): “…addition to the base classiﬁer-based system distributed with the platform, we train a variant which includes information from WordNet (Miller, 1995) and Verb-Ocean (Chklovski and Pantel, 2004), and makes use of features based on tree patterns and dependency tree skeletons (Wang and Neumann, 2007).”
From image descriptions to visual denotations: New similarity metrics for semantic inference over event descriptions (2014, methodology): “…(2008)’s balanced precision ( Bal ), which builds on Lin and on Weeds and Weir (2003)’s similarity ( W ): P We also use two publicly available resources that provide precomputed similarities, Kotlerman et al. (2010)’s D IRECT noun and verb rules and Chklovski and Pantel (2004)’s V ERB O CEAN rules.”
From Frequency to Meaning: Vector Space Models of Semantics (2010, methodology): “Relational classification: Chklovski and Pantel (2004) used a pair–pattern matrix to classify pairs of verbs into semantic classes.”
“Going on a vacation” takes longer than “Going for a walk”: A Study of Temporal Commonsense Understanding (2019, background): “…work has investigated some of these aspects, either explicitly or implicitly (e.g., duration (Gusev et al., 2011; Williams, 2012) and ordering (Chklovski and Pantel, 2004; Ning et al., 2018b)), but none of them have deﬁned or studied all aspects of temporal commonsense in a uniﬁed framework.”
SemEval-2010 Task 8: Multi-Way Classification of Semantic Relations Between Pairs of Nominals (2009, background): “More generally, semantic relation inventories attempt to generalize relations across wide groups of verbs (Chklovski and Pantel, 2004) and include relations that are not verbcentered (Nastase and Szpakowicz, 2003; Moldovan et al.”
Inherent Disagreements in Human Textual Inferences (2019, neutral): “In the second example, from DNC and derived from VerbNet (Chklovski and Pantel, 2004), there is disagreement about the manner aspect of ‘‘swat’’, that is, whether or not ‘‘swatting’’ is necessarily ‘‘forceful’’.”
Robust Textual Inference via Graph Matching (2005, background): “• LSA Match: v and M(v) are distributionally similar according to a freely available Latent Semantic Indexing package,2 or for verbs similar according to VerbOcean (Chklovski and Pantel, 2004).”
The Distributional Inclusion Hypotheses and Lexical Entailment (2005, background): “A few recent works (Ravichandran and Hovy, 2002; Keller et al., 2002; Chklovski and Pantel, 2004) used the web to collect statistics on word co-occurrences.”
A survey on opinion mining and sentiment analysis: Tasks, approaches and applications (2015, background): “[5], ConceptNet [120], VerbOcean [146], and ISEAR [147] to extract emotion expressing words.”
Recognizing textual entailment: Rational, evaluation and approaches – Erratum (2010, methodology): “Also DIRT (Lin and Pantel 2001), a resource of statistically learned inference rules, has been used in several systems, as well as verb-oriented resources such as VerbNet (Kipper-Schuler 2005) and VerbOcean (Chklovski and Pantel 2004).”
Learning Entailment Relations by Global Graph Structure Optimization (2012, unknown): “Chklovski and Pantel ( 2004 ) manually constructed a few dozen patterns and learned semantic relations between predicates by looking for these patterns on the Web.”
“Ask Not What Textual Entailment Can Do for You...” (2010, unknown): “An ablation study that formed part of the of-ﬁcial RTE 5 evaluation attempted to evaluate the contribution of publicly available knowledge resources such as WordNet (Fellbaum, 1998), VerbOcean (Chklovski and Pantel, 2004), and DIRT (Lin and Pantel, 2001) used by many of the systems.”
Machine Learning of Temporal Relations (2006, methodology): “We have taken pains to use freely download-able resources like Carafe, VerbOcean, and WEKA to help others easily replicate and quickly ramp up a system.”
Finding Contradictions in Text (2008, background): “Additionally, little external information is needed to gain broad coverage of antonymy, negation, and numeric mismatch contradictions; each involves only a closed set of words or data that can be obtained using existing resources and techniques (e.g., WordNet (Fellbaum, 1998), VerbOcean (Chklovski a”
Most “babies” are “little” and most “problems” are “huge”: Compositional Entailment in Adjective-Nouns (2016, methodology): “We test two systems from Excitement: the simple Maximum Entropy (MaxEnt) model which uses a suite of dense, similarity-based features (e.g. word overlap, cosine similarity), and the more sophisticated Maximum Entropy model (MaxEnt+LR) which uses the same similarity-based features but additionally in”
Combining Natural Logic and Shallow Reasoning for Question Answering (2016, methodology): “VERBOCEAN uses lexicosyntactic patterns to score pairs of verbs as candidate participants in a set of relations.”
Semantic Taxonomy Induction from Heterogenous Evidence (2006, background): “…variety of relationship-speciﬁc classiﬁers have been proposed, including pattern-based classiﬁers for hy-ponyms (Hearst, 1992), meronyms (Girju, 2003), synonyms (Lin et al., 2003), a variety of verb relations (Chklovski and Pantel, 2004), and general purpose analogy relations (Turney et al., 2003).”