Improving the Precision of Synset Links Between Cornetto and Princeton WordNet

摘要

Knowledge-based multilingual language processing benefits from having access to correctly es-tablished relations between semantic lexicons, such as the links between different WordNets.WordNet linking is a process that can be sped up by the use of computational techniques. Manualevaluations of the partly automatically established synonym set (synset) relations between Dutchand English in Cornetto, a Dutch lexical-semantic database associated with the EuroWordNetgrid, have confronted us with a worrisome amount of erroneous links. By extracting transla-tions from various bilingual resources and automatically assigning a confidence score to everypre-established link, we reduce the error rate of the existing equivalence relations between bothlanguages’ synsets (section 2). We will apply this technique to reuse the connection of Scleraand Beta pictograph sets and Cornetto synsets to Princeton WordNet and other WordNets, allow-ing us to further extend an existing Dutch text-to-pictograph translation tool to other languages(section 3).