Limits...
Harmony Search Algorithm for Word Sense Disambiguation.

Abed SA, Tiun S, Omar N - PLoS ONE (2015)

Bottom Line: Our proposed method was experimented on benchmark datasets, which yielded results comparable to the state-of-the-art WSD methods.In order to evaluate the effectiveness of the dependency generator, we perform the same methodology without the parser, but with a window of words.The empirical results demonstrate that the proposed method is able to produce effective solutions for most instances of the datasets used.

View Article: PubMed Central - PubMed

Affiliation: Knowledge Technology Research Group (KT), Centre for Artificial Intelligent (CAIT), Universiti Kebangsaan Malaysia, 43600 UKM Bangi, Selangor, Malaysia.

ABSTRACT
Word Sense Disambiguation (WSD) is the task of determining which sense of an ambiguous word (word with multiple meanings) is chosen in a particular use of that word, by considering its context. A sentence is considered ambiguous if it contains ambiguous word(s). Practically, any sentence that has been classified as ambiguous usually has multiple interpretations, but just one of them presents the correct interpretation. We propose an unsupervised method that exploits knowledge based approaches for word sense disambiguation using Harmony Search Algorithm (HSA) based on a Stanford dependencies generator (HSDG). The role of the dependency generator is to parse sentences to obtain their dependency relations. Whereas, the goal of using the HSA is to maximize the overall semantic similarity of the set of parsed words. HSA invokes a combination of semantic similarity and relatedness measurements, i.e., Jiang and Conrath (jcn) and an adapted Lesk algorithm, to perform the HSA fitness function. Our proposed method was experimented on benchmark datasets, which yielded results comparable to the state-of-the-art WSD methods. In order to evaluate the effectiveness of the dependency generator, we perform the same methodology without the parser, but with a window of words. The empirical results demonstrate that the proposed method is able to produce effective solutions for most instances of the datasets used.

No MeSH data available.


The collapsed form of the dependecy parses.“The dog scratched its back on the bark of the tree”
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4589330&req=5

pone.0136614.g003: The collapsed form of the dependecy parses.“The dog scratched its back on the bark of the tree”

Mentions: Obviously, the collapsed form of the dependencies shown in Fig 3 did shrink the number of stop words, such as prepositions words in the given example, with preservation of the grammatical structure of the sentence. Despite this, the collapsed form still carries useless grammatical relations in semantic measures, such as “det” and “root” in the given example. The proposed HSDG therefore neglects any of these useless relations; as they cannot be quantified semantically. However, the typed dependencies facilitate the task of disambiguating the sentence, where the ambiguous words in the sentence are disambiguated based on its heading or dependent word. This leads to excluding grammatically irrelevant words, which results in the noiseless substance of the sentence. For the purpose of explaining this process, we consider the word “bark” in the example shown in Fig 2. In the example, “bark” has two candidate senses based on the sentence context, which are “tough protective covering of the woody stems and roots of trees and other woody plants” and “the sound made by a dog”. The presence of the word “dog” in this example may mislead a WSD process in identifying the sense of “bark”. Hence, the significance of typed dependencies is obvious in directing the WSD system to choose the first sense of “bark” without confusion with the other sense.


Harmony Search Algorithm for Word Sense Disambiguation.

Abed SA, Tiun S, Omar N - PLoS ONE (2015)

The collapsed form of the dependecy parses.“The dog scratched its back on the bark of the tree”
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4589330&req=5

pone.0136614.g003: The collapsed form of the dependecy parses.“The dog scratched its back on the bark of the tree”
Mentions: Obviously, the collapsed form of the dependencies shown in Fig 3 did shrink the number of stop words, such as prepositions words in the given example, with preservation of the grammatical structure of the sentence. Despite this, the collapsed form still carries useless grammatical relations in semantic measures, such as “det” and “root” in the given example. The proposed HSDG therefore neglects any of these useless relations; as they cannot be quantified semantically. However, the typed dependencies facilitate the task of disambiguating the sentence, where the ambiguous words in the sentence are disambiguated based on its heading or dependent word. This leads to excluding grammatically irrelevant words, which results in the noiseless substance of the sentence. For the purpose of explaining this process, we consider the word “bark” in the example shown in Fig 2. In the example, “bark” has two candidate senses based on the sentence context, which are “tough protective covering of the woody stems and roots of trees and other woody plants” and “the sound made by a dog”. The presence of the word “dog” in this example may mislead a WSD process in identifying the sense of “bark”. Hence, the significance of typed dependencies is obvious in directing the WSD system to choose the first sense of “bark” without confusion with the other sense.

Bottom Line: Our proposed method was experimented on benchmark datasets, which yielded results comparable to the state-of-the-art WSD methods.In order to evaluate the effectiveness of the dependency generator, we perform the same methodology without the parser, but with a window of words.The empirical results demonstrate that the proposed method is able to produce effective solutions for most instances of the datasets used.

View Article: PubMed Central - PubMed

Affiliation: Knowledge Technology Research Group (KT), Centre for Artificial Intelligent (CAIT), Universiti Kebangsaan Malaysia, 43600 UKM Bangi, Selangor, Malaysia.

ABSTRACT
Word Sense Disambiguation (WSD) is the task of determining which sense of an ambiguous word (word with multiple meanings) is chosen in a particular use of that word, by considering its context. A sentence is considered ambiguous if it contains ambiguous word(s). Practically, any sentence that has been classified as ambiguous usually has multiple interpretations, but just one of them presents the correct interpretation. We propose an unsupervised method that exploits knowledge based approaches for word sense disambiguation using Harmony Search Algorithm (HSA) based on a Stanford dependencies generator (HSDG). The role of the dependency generator is to parse sentences to obtain their dependency relations. Whereas, the goal of using the HSA is to maximize the overall semantic similarity of the set of parsed words. HSA invokes a combination of semantic similarity and relatedness measurements, i.e., Jiang and Conrath (jcn) and an adapted Lesk algorithm, to perform the HSA fitness function. Our proposed method was experimented on benchmark datasets, which yielded results comparable to the state-of-the-art WSD methods. In order to evaluate the effectiveness of the dependency generator, we perform the same methodology without the parser, but with a window of words. The empirical results demonstrate that the proposed method is able to produce effective solutions for most instances of the datasets used.

No MeSH data available.