Metonymy Resolution as a Classification Task

  title={Metonymy Resolution as a Classification Task},
  author={Katja Markert and Malvina Nissim},
We reformulate metonymy resolution as a classification task. This is motivated by the regularity of metonymic readings and makes general classification and word sense disambiguation methods available for metonymy resolution. We then present a case study for location names, presenting both a corpus of location names annotated for metonymy as well as experiments with a supervised classification algorithm on this corpus. We especially explore the contribution of features used in word sense… 

Tables from this paper

Syntactic Features and Word Similarity for Supervised Metonymy Resolution

It is shown that syntactic head-modifier relations are a high precision feature for metonymy recognition but suffer from data sparseness, which is partially overcome by integrating a thesaurus and introducing simpler grammatical features, thereby preserving precision and increasing recall.

Data and models for metonymy resolution

The paper motivates the linguistic principles of data sampling and annotation and shows the task’s feasibility via human agreement and the first shared task for figurative language resolution, which was organised within SemEval-2007 and focused on metonymy.

XRCE-M: A Hybrid System for Named Entity Metonymy Resolution

A hybrid system based on the use of a robust parser that extracts deep syntactic relations combined with a non supervised distributional approach, also relying on the relations extracted by the parser.

Combining Collocations, Lexical and Encyclopedic Knowledge for Metonymy Resolution

A commonly used feature set with features extracted based on collocation information from corpora is enhanced using lexical and encyclopedic knowledge to determine the preferred sense of the potentially metonymic word using methods from unsupervised word sense disambiguation.

A Large Harvested Corpus of Location Metonymy

This work proposes a new, labelled, high-quality corpus of location metonymy called WiMCor, which is large in size and has high coverage, and uses different labels of varying granularity to annotate the corpus.

Whats in a Name? The Automatic Recognition of Metonymical Location Names

This paper investigates two approaches to metonymy recognition that dispense with this complexity, albeit in different ways, by replacing the complexity of current algorithms by a ‘lazy’ learning phase.

Metonymic Proper Names : A Corpus-based Account

A general framework for annotating metonymies in domainindependent text that considers the regularity, productivity and underspecification of metonymic usage is described and a gold standard corpus consisting of 4000 annotated occurrences of location and organisation names in the British National Corpus is presented.

Unsupervised approaches to metonymy recognition

Although the investigated technique, Schütze’s (1998) algorithm, enjoys considerable popularity in Word Sense Disambiguation, it is shown that it is not yet robust enough to tackle the specific case of metonymy recognition.

On metonymy recognition for geographic IR

Evaluation results indicate that, using a large annotated corpus of location names, a classifier based on shallow features achieves adequate performance, and removing metonymic senses from a database index yields a higher performance for GIR.

Impact of Target Word and Context on End-to-End Metonymy Detection

It is shown that the target word is less useful for detecting metonymy in the authors' dataset and the entity types that are associated with domain-specific words in their context are easier to solve, which shows that the context words are much more relevant for detecting meetonymy.



Towards a Corpus Annotated for Metonymies: the Case of Location Names

A framework for annotating metonymies in domain-independent text that considers the regularity, productivity and underspecification of metonymic usage is described and a fully worked out annotation scheme for location names is presented.

A Decision Tree of Bigrams is an Accurate Predictor of Word Sense

This paper presents a corpus-based approach to word sense disambiguation where a decision tree assigns a sense to an ambiguous word based on the bigrams that occur nearby. This approach is evaluated

Integrating Multiple Knowledge Sources to Disambiguate Word Sense: An Exemplar-Based Approach

This approach integrates a diverse set of knowledge sources to disambiguate word sense, including part of speech of neighboring words, morphological form, the unordered set of surrounding words, local collocations, and verb-object syntactic relation.

Unsupervised Word Sense Disambiguation Rivaling Supervised Methods

An unsupervised learning algorithm for sense disambiguation that, when trained on unannotated English text, rivals the performance of supervised techniques that require time-consuming hand annotations.

Understanding metonymies in discourse

A method for disambiguating word senses in a large corpus

The proposed method was designed to disambiguate senses that are usually associated with different topics using a Bayesian argument that has been applied successfully in related tasks such as author identification and information retrieval.

A Simple Approach to Building Ensembles of Naive Bayesian Classifiers for Word Sense Disambiguation

A corpus-based approach to word sense disambiguation that builds an ensemble of Naive Bayesian classifiers, each of which is based on lexical features that represent co-occurring words in varying sized windows of context achieves accuracy rivaling the best previously published results.

Two Kinds of Metonymy

It is shown that complications arise for both types of metonymy when multiple coercing predicates are considered and implemented algorithms handling these complexities are presented, as well as criteria for choosing one type of metonymic reading over another.

Deriving Metonymic Coercions from WordNet

The notion of metonymy is revisited and the computational aspects of its resolution in the context of the relational semantics provided by the recently released WordNet 1.6 lexical database is addressed.

A Corpus-based Account of Regular Polysemy: The Case of Context-sensitive Adjectives

A probabilistic model is proposed which provides a ranking on the set of possible interpretations of polysemous adjectives whose meaning varies depending on the nouns they modify and it is shown that the model's ranking of meanings correlates reliably with human intuitions.