Proceedings of the Morpho Challenge 2010 Workshop
No Thumbnail Available
Faculty of Information and Natural Sciences | C2 Toimitettu kirja, kokoomateos, konferenssijulkaisu tai lehden erikoisnumero
Unless otherwise stated, all rights belong to the author. You may download, display and print this publication for Your own personal use. Commercial use is prohibited.
TKK reports in information and computer science, 37
AbstractIn natural language processing many practical tasks, such as speech recognition, information retrieval and machine translation depend on a large vocabulary and statistical language models. For morphologically rich languages, such as Finnish and Turkish, the construction of a vocabulary and language models that have a sufficient coverage is particularly difficult, because of the huge amount of different word forms. In Morpho Challenge 2010 unsupervised and semi-supervised algorithms are suggested to provide morpheme analyses for words in different languages and evaluated in various practical applications. As a research theme, unsupervised morphological analysis has received wide attention in conferences and scientific journals focused on computational linguistic and its applications. This is the proceedings of the Morpho Challenge 2010 Workshop that contains one introduction article with a description of the tasks, evaluation and results and six articles describing the participating unsupervised and supervised learning algorithms. The Morpho Challenge 2010 Workshop was held at Espoo, Finland in 2-3 September, 2010.
morpheme analysis, unsupervised learning, semisupervised learning, information retrieval, statistical machine translation