The role of domain information in Word Sense Disambiguation

Magnini, Bernardo; Strapparava, Carlo; Pezzulo, Giovanni; Gliozzo, Alfio

doi:10.1017/s1351324902003029

Cited by 101 publications

(89 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Magnini et al [14] demonstrate that collapsing multiple senses to a set of senses belonging to a particular domain can address this problem. Although that exact solution is not relevant here, it would be profitable to investigate collapsing spurious senses in both the NLM training and testing data before applying the naïve Bayes classifier.…”

Section: Discussionmentioning

confidence: 99%

“…They tested their approach on the Senseval-2 dataset and evaluated whether directories were correctly assigned to words. Magnini et al [14] also used WordNet for the Senseval-2 dataset but extended it by adding domain names such as Medicine or Architecture to every synset. They assigned these to a subset of words in the text based on frequencies of the domains and a few additional rules.…”

Section: Information Sourcesmentioning

confidence: 99%

See 1 more Smart Citation

Effects of information and machine learning algorithms on word sense disambiguation with small datasets

Leroy

Rindflesch

2005

International Journal of Medical Informatics

View full text Add to dashboard Cite

Summary Current approaches to word sense disambiguation use (and often combine) various machine learning techniques. Most refer to characteristics of the ambiguity and its surrounding words and are based on thousands of examples. Unfortunately, developing large training sets is burdensome, and in response to this challenge, we investigate the use of symbolic knowledge for small datasets. A naïve Bayes classifier was trained for 15 words with 100 examples for each. Unified Medical Language System (UMLS) semantic types assigned to concepts found in the sentence and relationships between these semantic types form the knowledge base. The most frequent sense of a word served as the baseline. The effect of increasingly accurate symbolic knowledge was evaluated in nine experimental conditions. Performance was measured by accuracy based on 10-fold cross-validation. The best condition used only the semantic types of the words in the sentence. Accuracy was then on average 10% higher than the baseline; however, it varied from 8% deterioration to 29% improvement. To investigate this large variance, we performed several followup evaluations, testing additional algorithms (decision tree and neural network), and gold standards (per expert), but the results did not significantly differ. However, we noted a trend that the best disambiguation was found for words that were the least troublesome to the human evaluators. We conclude that neither algorithm nor individual human behavior cause these large differences, but that the structure of the UMLS Metathesaurus (used to represent senses of ambiguous words) contributes to inaccuracies in the gold standard, leading to varied performance of word sense disambiguation techniques.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Information Sourcesmentioning

confidence: 99%

Effects of information and machine learning algorithms on word sense disambiguation with small datasets

Leroy

Rindflesch

2005

International Journal of Medical Informatics

View full text Add to dashboard Cite

show abstract

“…Additionally, we evaluated our method quantitatively using the Subject Field Codes (SFC) resource (Magnini and Cavaglià, 2000) which annotates WordNet synsets with domain labels. The SFC contains an economy label and a sports label.…”

Section: Two Experimentsmentioning

confidence: 99%

Finding predominant word senses in untagged text

McCarthy

Koeling

Weeds

et al. 2004

Proceedings of the 42nd Annual Meeting on Association for Computational Linguistics - ACL '04

193

233

View full text Add to dashboard Cite

In word sense disambiguation (WSD), the heuristic of choosing the most common sense is extremely powerful because the distribution of the senses of a word is often skewed. The problem with using the predominant, or first sense heuristic, aside from the fact that it does not take surrounding context into account, is that it assumes some quantity of handtagged data. Whilst there are a few hand-tagged corpora available for some languages, one would expect the frequency distribution of the senses of words, particularly topical words, to depend on the genre and domain of the text under consideration. We present work on the use of a thesaurus acquired from raw textual corpora and the WordNet similarity package to find predominant noun senses automatically. The acquired predominant senses give a precision of 64% on the nouns of the SENSEVAL-2 English all-words task. This is a very promising result given that our method does not require any hand-tagged text, such as SemCor. Furthermore, we demonstrate that our method discovers appropriate predominant senses for words from two domainspecific corpora.

show abstract

“…This domain relatedness (or lack thereof) was successfully used in the past for word sense disambiguation [5,11] and also for the generation of jokes [21]. We thus conduct experiments to check whether domain similarity and/or opposition can constitute a feature to discriminate the humorous punch line.…”

Section: Domain Fitnessmentioning

confidence: 99%

“…WORDNET DOMAINS organizes about 250 domain labels in a hierarchy, exploiting Dewey Decimal Classification. Following [11], we consider an intermediate level of the domain hierarchy, consisting of 42 disjoint labels (i.e. we use SPORT instead of VOLLEY or BASKETBALL, which are subsumed by SPORT).…”

Section: Domain Fitnessmentioning

confidence: 99%

Computational Models for Incongruity Detection in Humour

Mihalcea¹,

Strapparava²,

Pulman³

2010

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

Abstract. Incongruity resolution is one of the most widely accepted theories of humour, suggesting that humour is due to the mixing of two disparate interpretation frames in one statement. In this paper, we explore several computational models for incongruity resolution. We introduce a new data set, consisting of a series of 'set-ups' (preparations for a punch line), each of them followed by four possible coherent continuations out of which only one has a comic effect. Using this data set, we redefine the task as the automatic identification of the humorous punch line among all the plausible endings. We explore several measures of semantic relatedness, along with a number of joke-specific features, and try to understand their appropriateness as computational models for incongruity detection.

show abstract

The role of domain information in Word Sense Disambiguation

Cited by 101 publications

References 8 publications

Effects of information and machine learning algorithms on word sense disambiguation with small datasets

Effects of information and machine learning algorithms on word sense disambiguation with small datasets

Finding predominant word senses in untagged text

Computational Models for Incongruity Detection in Humour

Contact Info

Product

Resources

About