Difference between revisions of "Meliorated Approach for Extracting Bilingual Terminology from Wikipedia"

From Wikipedia Quality
Jump to: navigation, search
(+ links)
(Infobox)
 
Line 1: Line 1:
 +
{{Infobox work
 +
| title = Meliorated Approach for Extracting Bilingual Terminology from Wikipedia
 +
| date = 2008
 +
| authors = [[Anand Gupta]]<br />[[Akhil Goyal]]<br />[[Aman Bindal]]<br />[[Ankuj Gupta]]
 +
| doi = 10.1109/ICCITECHN.2008.4803031
 +
| link = http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=4803031
 +
}}
 
'''Meliorated Approach for Extracting Bilingual Terminology from Wikipedia''' - scientific work related to [[Wikipedia quality]] published in 2008, written by [[Anand Gupta]], [[Akhil Goyal]], [[Aman Bindal]] and [[Ankuj Gupta]].
 
'''Meliorated Approach for Extracting Bilingual Terminology from Wikipedia''' - scientific work related to [[Wikipedia quality]] published in 2008, written by [[Anand Gupta]], [[Akhil Goyal]], [[Aman Bindal]] and [[Ankuj Gupta]].
  
 
== Overview ==
 
== Overview ==
 
With the demand of accurate and domain specific bilingual dictionaries, research in the field of automatic dictionary extraction has become popular. Due to lack of domain specific terminology in parallel corpora, extraction of bilingual terminology from [[Wikipedia]] (a corpus for knowledge extraction having a huge amount of articles, links within [[different language]]s, a dense link structure and a number of redirect pages) has taken up a new research in the field of bilingual dictionary creation. Authors method not only analyzes interlanguage links along with redirect page titles and link text titles but also filters out inaccurate translation candidates using pattern matching. Score of each translation candidate is calculated using page parameters and then setting an appropriate threshold as compared to previous approach, which was solely, based on backward links. In experiment, authors proved the advantages of approach compared to the traditional approach.
 
With the demand of accurate and domain specific bilingual dictionaries, research in the field of automatic dictionary extraction has become popular. Due to lack of domain specific terminology in parallel corpora, extraction of bilingual terminology from [[Wikipedia]] (a corpus for knowledge extraction having a huge amount of articles, links within [[different language]]s, a dense link structure and a number of redirect pages) has taken up a new research in the field of bilingual dictionary creation. Authors method not only analyzes interlanguage links along with redirect page titles and link text titles but also filters out inaccurate translation candidates using pattern matching. Score of each translation candidate is calculated using page parameters and then setting an appropriate threshold as compared to previous approach, which was solely, based on backward links. In experiment, authors proved the advantages of approach compared to the traditional approach.

Latest revision as of 02:28, 24 May 2020


Meliorated Approach for Extracting Bilingual Terminology from Wikipedia
Authors
Anand Gupta
Akhil Goyal
Aman Bindal
Ankuj Gupta
Publication date
2008
DOI
10.1109/ICCITECHN.2008.4803031
Links
Original

Meliorated Approach for Extracting Bilingual Terminology from Wikipedia - scientific work related to Wikipedia quality published in 2008, written by Anand Gupta, Akhil Goyal, Aman Bindal and Ankuj Gupta.

Overview

With the demand of accurate and domain specific bilingual dictionaries, research in the field of automatic dictionary extraction has become popular. Due to lack of domain specific terminology in parallel corpora, extraction of bilingual terminology from Wikipedia (a corpus for knowledge extraction having a huge amount of articles, links within different languages, a dense link structure and a number of redirect pages) has taken up a new research in the field of bilingual dictionary creation. Authors method not only analyzes interlanguage links along with redirect page titles and link text titles but also filters out inaccurate translation candidates using pattern matching. Score of each translation candidate is calculated using page parameters and then setting an appropriate threshold as compared to previous approach, which was solely, based on backward links. In experiment, authors proved the advantages of approach compared to the traditional approach.