Word Sense Disambiguation based on Semantic Density Page: 1
7 p.View a full description of this paper.
Extracted Text
The following text was automatically extracted from the image on this page using optical character recognition software:
Word Sense Disambiguation based on Semantic Density
Rada Mihalcea and Dan I. Moldovan
Department of Computer Science and Engineering
Southern Methodist University
Dallas, Texas, 75275-0122
{rada,moldovan}@seas .smu.edu
Abstract
This paper presents a Word Sense Disambiguation method based on the idea of semantic density
between words. The disambiguation is done in the context of WordNet. The Internet is used as a
raw corpora to provide statistical information for word associations. A metric is introduced and
used to measure the semantic density and to rank all possible combinations of the senses of two
words. This method provides a precision of 58% in indicating the correct sense for both words at
the same time. The precision increases as we consider more choices: 70% for top two ranked and
73% for top three ranked.1 Introduction
Word Sense Disambiguation (WSD) is an open prob-
lem in Natural Language Processing. Its solution
impacts other tasks such as discourse, reference reso-
lution, coherence, inference and others. WSD meth-
ods can be broadly classified into three types:
1. WSD that make use of the information pro-
vided by machine readable dictionaries (Cowie
et al.1992), (Miller et al.1994), (Agirre and
Rigau, 1995), (Li et al.1995), (McRoy, 1992);
2. WSD that use information gathered from train-
ing on a corpus that has already been semanti-
cally disambiguated (supervised training meth-
ods) (Gale, Church et al., 1992), (Ng and Lee,
1996);
3. WSD that use information gathered from
raw corpora (unsupervised training methods)
(Yarowsky 1995) (Resnik 1997).
There are also hybrid methods that combine sev-
eral sources of knowledge such as lexicon informa-
tion, heuristics, collocations and others (McRoy,
1992) (Bruce and Wiebe, 1994) (Ng and Lee, 1996)
(Rigau, Asterias et al., 1997).
Statistical methods produce high accuracy results
for small number of preselected words. A lack of
widely available semantically tagged corpora almost
excludes supervised learning methods. On the other
hand, the disambiguation using unsupervised meth-
ods has the disadvantage that the senses are not well
defined. To our knowledge, none of the statistical
methods disambiguate adjectives or adverbs so far.
One approach to WSD is to determine the concep-
tual distance between words, that is to measure the
semantic closeness of the words within a semantic
network. Essentially, it is the length of the short-
est path connecting the concepts (Rada et al.1989),
(Rigau, Asterias et al., 1997). By measuring theconceptual distance between words, it is possible to
determine the likelihood of word sense associations.
For example, the method proposed in (Li et al.1995)
tries to determine the possible sense of a noun asso-
ciated with a verb using WordNet and a large text.
Based on other occurrences of the verb or semanti-
cally related verbs in the text, the possible object
is determined by measuring the semantic similarity
between the noun objects.
Methods that do not need large corpora are usu-
ally based exclusively on MRD. A proposal in this
sense has been made in (Agirre and Rigau, 1995);
they measure the conceptual density between nouns,
by using WordNet, but the method proposed in their
paper cannot be applied to measuring a concep-
tual distance between a verb and a noun, as no di-
rect links are provided in MRDs between the nouns
and verbs hierarchies. A WordNet-based method
for measuring the semantic similarity between nouns
was also proposed in (Richardson et al., 1994). Their
method consists of using hierarchical concept graphs
constructed from WordNet data files, and a semantic
similarity formula. Still, the method does not pro-
vide a link between different part-of-speech words.
2 Our approach
The approach described in this paper is based on the
idea of semantic density. This can be measured by
the number of common words that are within a se-
mantic distance of two or more words. The closer the
semantic relationship between two words the higher
the semantic density between them. The way it is
defined here, the semantic density works well in the
case of uniform MRD. In reality there are gaps in the
knowledge representations and the semantic density
can provide only an estimation of the actual seman-
tic relatedness between words.
We introduce the semantic density because it is
Upcoming Pages
Here’s what’s next.
Search Inside
This paper can be searched. Note: Results may vary based on the legibility of text within the document.
Tools / Downloads
Get a copy of this page or view the extracted text.
Citing and Sharing
Basic information for referencing this web page. We also provide extended guidance on usage rights, references, copying or embedding.
Reference the current page of this Paper.
Mihalcea, Rada, 1974- & Moldovan, Dan I. Word Sense Disambiguation based on Semantic Density, paper, August 1998; (https://digital.library.unt.edu/ark:/67531/metadc83303/m1/1/: accessed April 25, 2024), University of North Texas Libraries, UNT Digital Library, https://digital.library.unt.edu; crediting UNT College of Engineering.