Magnus Merkel (Department of Computer and Information Science

Linköping University and Fodina Language Technology AB) talar under rubriken "Term extraction and building semantic relations"


Tid och plats: Torsdagen den 5 november 2009, kl. 15-17 i C 307


In this talk I will present the work done at Linköping on multilingual term extraction, i.e. using word alignment to extract links between words and phrases from parallel corpora, extract term candidates, and use semantic mirror-like techniques to establish synonym sets and hyponym relations between term sets. I will also describe the process of going from a parallel patent document set to a term database consisting of hierarchically validated term translations of English-Swedish patent terms.