CITI has stopped operations in 2014, to co-launch NOVA LINCS THIS SITE IS NOT BEING UPDATED SINCE 2013
citi banner
  Home  \  Publications  \  InProceedings Page Login  
banner bottom
File Top
Context Sense Clustering for Translation

Extended Abstract Word sense ambiguity is present in all words with more than one meaning in several natural languages and is a fundamental characteristic of human language. This has consequences in translation as it is necessary to find the right sense and the correct translation for each word. For this reason, the English word "fair" can mean "reasonable" or "market" such as "plant" also can mean "factory" or "herb". The disambiguation problem has been recognize as a major problem in natural languages processing research. Several words have several meanings or senses. The disambiguation task seeks to find out which sense of an ambiguous word is invoked in a particular use of that word. A system for automatic translation from English to Portuguese should know how to translate the word "bank" as "banco" (an institution for receiving, lending, exchanging, and safeguarding money), and as "margem" (the land alongside or sloping down to a river or lake), and also should know that the word "banana" may appear in the same context as "acerola" and that these two belongs to hyperonym "fruit". Whenever a translation systems depends on the meaning of the text being processed, disambiguation is beneficial or even necessary. Word Sense Disambiguation is thus essentially a classification problem; given a word X and an inventory of possible semantic tags for that word that might be translation, we seek which tag is appropriate for each individual instance of that word in a particularly context. ...


@ Proceedings of SSST-8, Eighth Workshop on Syntax, Semantics and Structure in Statistical Translation

Series: EMNLP Conference, Computational Linguistics Association

Publisher: Association for Computational Linguistics ( United States )

Pages: 135 to 137

Date: October, 2014


File Bottom