Compound splitting: Difference between revisions

From Clarin K-Centre
Jump to navigation Jump to search
No edit summary
No edit summary
 
Line 9: Line 9:
*[https://biblio.ugent.be/publication/7126122 Lieve Macken and Arda Tezcan. 2018. “Dutch Compound Splitting for Bilingual Terminology Extraction.” In Multiword Units in Machine Translation and Translation Technology, ed. Ruslan Mitkov, Johanna Monti, Gloria Corpas Pastor, and Violeta Seretan. Vol. 341. John Benjamins, pp. 148–162.]
*[https://biblio.ugent.be/publication/7126122 Lieve Macken and Arda Tezcan. 2018. “Dutch Compound Splitting for Bilingual Terminology Extraction.” In Multiword Units in Machine Translation and Translation Technology, ed. Ruslan Mitkov, Johanna Monti, Gloria Corpas Pastor, and Violeta Seretan. Vol. 341. John Benjamins, pp. 148–162.]
*[https://lt3.ugent.be/compound-splitter-demo/ Demo]
*[https://lt3.ugent.be/compound-splitter-demo/ Demo]
==CharSplit - An ngram-based compound splitter==
Python module that splits a compound into its body and head. So far German and Dutch are supported.
*[https://pypi.org/project/compound-split/ Webpage]
==Wordbuilder==
*[https://www.aclweb.org/anthology/L02-1004/ Vincent Vandeghinste (2002). Lexicon Optimization: Maximizing Lexical Coverage in Speech Recognition through Automated Compounding.] Proceedings of the Third International Conference on Language Resources and Evaluation (LREC2002). ELRA. Paris.
</translate>
</translate>

Latest revision as of 15:29, 6 November 2024

Other languages:

Compound splitter demo

A compound splitter splits compounds into their component parts, e.g. liefde+s+drank or [post+zegel]+verzamelaar. This demo allows Dutch input up to 500 characters. You can either input running text or single words (one word per line). If you are interested in using the compound splitter for other purposes contact Lieve.Macken@UGent.be.

CharSplit - An ngram-based compound splitter

Python module that splits a compound into its body and head. So far German and Dutch are supported.

Wordbuilder