Morpho challenge: evaluation of algorithms for unsupervised learning of morphology in various tasks and languages

Authors:
Mikko Kurimo;Sami Virpioja;Ville Turunen;Teemu Hirsimäki
Affiliations:
Helsinki University of Technology, TKK, Finland;Helsinki University of Technology, TKK, Finland;Helsinki University of Technology, TKK, Finland;Helsinki University of Technology, TKK, Finland
Venue:
NAACL-Demonstrations '09 Proceedings of Human Language Technologies: The 2009 Annual Conference of the North American Chapter of the Association for Computational Linguistics, Companion Volume: Demonstration Session
Year:
2009

Citing 3
Cited 0

Unsupervised discovery of morphemes

MPL '02 Proceedings of the ACL-02 workshop on Morphological and phonological learning - Volume 6
Unlimited vocabulary speech recognition for agglutinative languages

HLT-NAACL '06 Proceedings of the main conference on Human Language Technology Conference of the North American Chapter of the Association of Computational Linguistics
Minimum Bayes risk combination of translation hypotheses from alternative morphological decompositions

NAACL-Short '09 Proceedings of Human Language Technologies: The 2009 Annual Conference of the North American Chapter of the Association for Computational Linguistics, Companion Volume: Short Papers

Quantified Score

Hi-index	0.00

Visualization

Abstract

After the release of the open source software implementation of Morfessor algorithm, a series of several open evaluations has been organized for unsupervised morpheme analysis and morpheme-based speech recognition and information retrieval. The unsupervised morpheme analysis is a particularly attractive approach for speech and language technology for the morphologically complex languages. When the amount of distinct word forms becomes prohibitive for the construction of a sufficient lexicon, it is important that the words can be segmented into smaller meaningful language modeling units. In this presentation we will demonstrate the results of the evaluations, the baseline systems built using the open source tools, and invite research groups to participate in the next evaluation where the task is to enhance statistical machine translation by morpheme analysis.