Word Sense Disambiguation Improves Information Retrieval
Zhong, Zhi and Ng, Hwee Tou

Article Structure

Abstract

Previous research has conflicting conclusions on whether word sense disambiguation (WSD) systems can improve information retrieval (IR) performance.

Introduction

Word sense disambiguation (WSD) is the task of identifying the correct meaning of a word in context.

Related Work

Many previous studies have analyzed the benefits and the problems of applying WSD to IR.

The Language Modeling Approach to IR

This section describes the LM approach to IR and the pseudo relevance feedback approach.

Word Sense Disambiguation

In this section, we first describe the construction of our WSD system.

Incorporating Senses into Language Modeling Approaches

In this section, we propose to incorporate senses into the LM approach to IR.

Experiments

In this section, we evaluate and analyze the models proposed in Section 5 on standard TREC collections.

Conclusion

This paper reports successful application of WSD to IR.

Topics

word senses

Appears in 13 sentences as: Word sense (2) word sense (2) word senses (10)
In Word Sense Disambiguation Improves Information Retrieval
  1. Previous research has conflicting conclusions on whether word sense disambiguation (WSD) systems can improve information retrieval (IR) performance.
    Page 1, “Abstract”
  2. Together with the senses predicted for words in documents, we propose a novel approach to incorporate word senses into the language modeling approach to IR and also exploit the integration of synonym relations.
    Page 1, “Abstract”
  3. Our experimental results on standard TRE C collections show that using the word senses tagged by a supervised WSD system, we obtain significant improvements over a state-of-the-art IR system.
    Page 1, “Abstract”
  4. Word sense disambiguation (WSD) is the task of identifying the correct meaning of a word in context.
    Page 1, “Introduction”
  5. Some of the early research showed a drop in retrieval performance by using word senses (Krovetz and Croft, 1992; Voorhees, 1993).
    Page 1, “Introduction”
  6. Some other experiments observed improvements by integrating word senses in IR systems (Schutze and Pedersen, 1995; Gonzalo et al., 1998; Stokoe et al., 2003; Kim et al., 2004).
    Page 1, “Introduction”
  7. This paper proposes the use of word senses to improve the performance of IR.
    Page 1, “Introduction”
  8. We incorporate word senses into the language modeling (LM) approach to IR (Ponte and Croft, 1998), and utilize sense synonym relations to further improve the performance.
    Page 1, “Introduction”
  9. generating word senses for query terms in Section 4, followed by presenting our novel method of incorporating word senses and their synonyms into the LM approach in Section 5.
    Page 2, “Introduction”
  10. However, it is hard to judge the effect of word senses because of the overall poor performances of their baseline method and their system.
    Page 2, “Related Work”
  11. 4.1 Word sense disambiguation system
    Page 4, “Word Sense Disambiguation”

See all papers in Proc. ACL 2012 that mention word senses.

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

Back to top.

LM

Appears in 11 sentences as: LM (11)
In Word Sense Disambiguation Improves Information Retrieval
  1. We incorporate word senses into the language modeling ( LM ) approach to IR (Ponte and Croft, 1998), and utilize sense synonym relations to further improve the performance.
    Page 1, “Introduction”
  2. Section 3 introduces the LM approach to IR, including the pseudo relevance feedback method.
    Page 1, “Introduction”
  3. generating word senses for query terms in Section 4, followed by presenting our novel method of incorporating word senses and their synonyms into the LM approach in Section 5.
    Page 2, “Introduction”
  4. This section describes the LM approach to IR and the pseudo relevance feedback approach.
    Page 3, “The Language Modeling Approach to IR”
  5. Suppose we already have a basic IR method which does not require any sense information, such as the stem-based LM approach.
    Page 4, “Word Sense Disambiguation”
  6. In this section, we propose to incorporate senses into the LM approach to IR.
    Page 5, “Incorporating Senses into Language Modeling Approaches”
  7. In this part, we further integrate the synonym relations of senses into the LM approach.
    Page 5, “Incorporating Senses into Language Modeling Approaches”
  8. We use the Lemur toolkit (Ogilvie and Callan, 2001) version 4.11 as the basic retrieval tool, and select the default unigram LM approach based on KL-divergence and Dirichlet-prior smoothing method in Lemur as our basic retrieval approach.
    Page 6, “Experiments”
  9. We proposed a method for annotating senses to terms in short queries, and also described an approach to integrate senses into an LM approach for IR.
    Page 8, “Conclusion”
  10. Our experimental results showed that the incorporation of senses improved a state-of-the-art baseline, a stem-based LM approach with PRF method.
    Page 8, “Conclusion”
  11. We also proposed a method to further integrate the synonym relations to the LM approaches.
    Page 8, “Conclusion”

See all papers in Proc. ACL 2012 that mention LM.

See all papers in Proc. ACL that mention LM.

Back to top.

language model

Appears in 10 sentences as: language model (5) language modeling (5) language models (3)
In Word Sense Disambiguation Improves Information Retrieval
  1. Together with the senses predicted for words in documents, we propose a novel approach to incorporate word senses into the language modeling approach to IR and also exploit the integration of synonym relations.
    Page 1, “Abstract”
  2. We incorporate word senses into the language modeling (LM) approach to IR (Ponte and Croft, 1998), and utilize sense synonym relations to further improve the performance.
    Page 1, “Introduction”
  3. 3.1 The language modeling approach
    Page 3, “The Language Modeling Approach to IR”
  4. In the language modeling approach to IR, language models are constructed for each query (1 and each document d in a text collection C. The documents in C are ranked by the distance to a given query (1 according to the language models .
    Page 3, “The Language Modeling Approach to IR”
  5. The most commonly used language model in IR is the unigram model, in which terms are assumed to be independent of each other.
    Page 3, “The Language Modeling Approach to IR”
  6. In the rest of this paper, language model will refer to the unigram language model .
    Page 3, “The Language Modeling Approach to IR”
  7. In the calculation of p(t|6d), several smoothing methods have been proposed to overcome the data sparseness problem of a language model constructed from one document (Zhai and Lafferty, 2001b).
    Page 3, “The Language Modeling Approach to IR”
  8. The next problem is to incorporate the sense information into the language modeling approach.
    Page 5, “Incorporating Senses into Language Modeling Approaches”
  9. Given a query (1 and a document d in text collection C, we want to reestimate the language models by making use of the sense information assigned to them.
    Page 5, “Incorporating Senses into Language Modeling Approaches”
  10. With this language model , the probability of a query term in a document is enlarged by the synonyms of its senses; The more its synonym senses in a document, the higher the probability.
    Page 6, “Incorporating Senses into Language Modeling Approaches”

See all papers in Proc. ACL 2012 that mention language model.

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

Back to top.

significant improvements

Appears in 6 sentences as: significant improvements (4) significantly improves (2)
In Word Sense Disambiguation Improves Information Retrieval
  1. Our experimental results on standard TRE C collections show that using the word senses tagged by a supervised WSD system, we obtain significant improvements over a state-of-the-art IR system.
    Page 1, “Abstract”
  2. In the application of WSD to MT, research has shown that integrating WSD in appropriate ways significantly improves the performance of MT systems (Chan et al., 2007; Carpuat and Wu, 2007).
    Page 1, “Introduction”
  3. Our evaluation on standard TREC1 data sets shows that supervised WSD outperforms two other WSD baselines and significantly improves IR.
    Page 1, “Introduction”
  4. They obtained significant improvements by representing documents and queries with accurate senses as well as synsets (synonym sets).
    Page 2, “Related Work”
  5. Their evaluation on TREC collections achieved significant improvements over a standard term based vector space model.
    Page 2, “Related Work”
  6. tistically significant improvements over Stemprf on TREC7, TREC8, and RBO3.
    Page 8, “Experiments”

See all papers in Proc. ACL 2012 that mention significant improvements.

See all papers in Proc. ACL that mention significant improvements.

Back to top.

semantic relations

Appears in 4 sentences as: semantic relations (4)
In Word Sense Disambiguation Improves Information Retrieval
  1. The utilization of semantic relations has proved to be helpful for IR.
    Page 2, “Related Work”
  2. ing to investigate the utilization of semantic relations among senses in IR.
    Page 3, “Related Work”
  3. Words usually have some semantic relations with others.
    Page 5, “Incorporating Senses into Language Modeling Approaches”
  4. Synonym relation is one of the semantic relations commonly used to improve IR performance.
    Page 5, “Incorporating Senses into Language Modeling Approaches”

See all papers in Proc. ACL 2012 that mention semantic relations.

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

Back to top.

unigram

Appears in 4 sentences as: unigram (4)
In Word Sense Disambiguation Improves Information Retrieval
  1. The most commonly used language model in IR is the unigram model, in which terms are assumed to be independent of each other.
    Page 3, “The Language Modeling Approach to IR”
  2. In the rest of this paper, language model will refer to the unigram language model.
    Page 3, “The Language Modeling Approach to IR”
  3. With unigram model, the negative KL-divergence between model 6.1 of query (1 and model 6d of document d is calculated as follows:
    Page 3, “The Language Modeling Approach to IR”
  4. We use the Lemur toolkit (Ogilvie and Callan, 2001) version 4.11 as the basic retrieval tool, and select the default unigram LM approach based on KL-divergence and Dirichlet-prior smoothing method in Lemur as our basic retrieval approach.
    Page 6, “Experiments”

See all papers in Proc. ACL 2012 that mention unigram.

See all papers in Proc. ACL that mention unigram.

Back to top.

WordNet

Appears in 4 sentences as: WordNet (4)
In Word Sense Disambiguation Improves Information Retrieval
  1. Voorhees (1993) used the hyponymy (“ISA”) relation in WordNet (Miller, 1990) to disambiguate the polysemous nouns in a text.
    Page 2, “Related Work”
  2. (2004) tagged words with 25 root senses of nouns in WordNet .
    Page 2, “Related Work”
  3. Some researchers achieved improvements by expanding the disambiguated query words with synonyms and some other information from WordNet (Voorhees, 1994; Liu et al., 2004; Liu et al., 2005; Fang, 2008).
    Page 2, “Related Work”
  4. The usage of knowledge sources from WordNet in document expansion also showed improvements in IR systems (Cao et al., 2005; Agirre et al., 2010).
    Page 2, “Related Work”

See all papers in Proc. ACL 2012 that mention WordNet.

See all papers in Proc. ACL that mention WordNet.

Back to top.

parallel corpora

Appears in 3 sentences as: parallel corpora (3)
In Word Sense Disambiguation Improves Information Retrieval
  1. We construct our supervised WSD system directly from parallel corpora .
    Page 4, “Word Sense Disambiguation”
  2. To generate the WSD training data, 7 parallel corpora were used, including Chinese Treebank, FBIS Corpus, Hong Kong Hansards, Hong Kong Laws, Hong Kong News, Sinorama News Magazine, and Xinhua Newswire.
    Page 4, “Word Sense Disambiguation”
  3. Then, word alignment was performed on the parallel corpora with the GIZA+ + software (Och and Ney, 2003).
    Page 4, “Word Sense Disambiguation”

See all papers in Proc. ACL 2012 that mention parallel corpora.

See all papers in Proc. ACL that mention parallel corpora.

Back to top.

sense disambiguation

Appears in 3 sentences as: sense disambiguation (3)
In Word Sense Disambiguation Improves Information Retrieval
  1. Previous research has conflicting conclusions on whether word sense disambiguation (WSD) systems can improve information retrieval (IR) performance.
    Page 1, “Abstract”
  2. Word sense disambiguation (WSD) is the task of identifying the correct meaning of a word in context.
    Page 1, “Introduction”
  3. 4.1 Word sense disambiguation system
    Page 4, “Word Sense Disambiguation”

See all papers in Proc. ACL 2012 that mention sense disambiguation.

See all papers in Proc. ACL that mention sense disambiguation.

Back to top.

word alignment

Appears in 3 sentences as: word aligned (1) word alignment (2)
In Word Sense Disambiguation Improves Information Retrieval
  1. Then, word alignment was performed on the parallel corpora with the GIZA+ + software (Och and Ney, 2003).
    Page 4, “Word Sense Disambiguation”
  2. For each English morphological root 6, the English sentences containing its occurrences were eXtracted from the word aligned output of GIZA++, as well as the corresponding translations of these occurrences.
    Page 4, “Word Sense Disambiguation”
  3. To minimize noisy word alignment result, translations with no Chinese character were deleted, and we further removed a translation when it only appears once, or its frequency is less than 10 and also less than 1% of the frequency of 6.
    Page 4, “Word Sense Disambiguation”

See all papers in Proc. ACL 2012 that mention word alignment.

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

Back to top.

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