Automatic enhancement of LTAG Treebanks
Research output: Contribution to journal › Conference article › Research › peer-review
Standard
Automatic enhancement of LTAG Treebanks. / Zarei, Farzaneh; Basirat, Ali; Faili, Heshaam; Mirian, Maryam Sadat.
In: International Conference Recent Advances in Natural Language Processing, RANLP, 2013, p. 733-739.Research output: Contribution to journal › Conference article › Research › peer-review
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - GEN
T1 - Automatic enhancement of LTAG Treebanks
AU - Zarei, Farzaneh
AU - Basirat, Ali
AU - Faili, Heshaam
AU - Mirian, Maryam Sadat
PY - 2013
Y1 - 2013
N2 - The Treebanks as the sets of syntactically annotated sentences, are the most widely used language resource in the application of Natural Language Processing. The occurrence of errors in the automatically created Treebanks is one of the main obstacles limiting the using of these resources in the real world applications. This paper aims to introduce an statistical method for diminishing the amount of errors occurred in a specific English LTAG-Treebank proposed in Basirat and Faili (2013). The problem has been formulated as a classification problem and has been tackled by using several classifiers. The experiments show that by using this approach, about 95% of the errors could be detected and more than 77% of them could successfully be corrected in the case of using Adaboost classifier. In addition, it has been shown that the new treebank could reach a high of 76% F-measure which is 8% higher than the original treebank.
AB - The Treebanks as the sets of syntactically annotated sentences, are the most widely used language resource in the application of Natural Language Processing. The occurrence of errors in the automatically created Treebanks is one of the main obstacles limiting the using of these resources in the real world applications. This paper aims to introduce an statistical method for diminishing the amount of errors occurred in a specific English LTAG-Treebank proposed in Basirat and Faili (2013). The problem has been formulated as a classification problem and has been tackled by using several classifiers. The experiments show that by using this approach, about 95% of the errors could be detected and more than 77% of them could successfully be corrected in the case of using Adaboost classifier. In addition, it has been shown that the new treebank could reach a high of 76% F-measure which is 8% higher than the original treebank.
UR - http://www.scopus.com/inward/record.url?scp=84890452027&partnerID=8YFLogxK
M3 - Conference article
AN - SCOPUS:84890452027
SP - 733
EP - 739
JO - International Conference Recent Advances in Natural Language Processing, RANLP
JF - International Conference Recent Advances in Natural Language Processing, RANLP
SN - 1313-8502
T2 - 9th International Conference on Recent Advances in Natural Language Processing, RANLP 2013
Y2 - 9 September 2013 through 11 September 2013
ER -
ID: 366047604