Deciphering Undersegmented Ancient Scripts Using Phonetic Prior

Cite This

Files in this item

Checksum: MD5:5926f4267ae46c209442167af6544897

LUO, Jiaming, Frederik HARTMANN, Enrico SANTUS, Regina BARZILAY, Yuan CAO, 2021. Deciphering Undersegmented Ancient Scripts Using Phonetic Prior. In: Transactions of the Association for Computational Linguistics. MIT Press. 9, pp. 69-81. eISSN 2307-387X. Available under: doi: 10.1162/tacl_a_00354

@article{Luo2021Decip-57257, title={Deciphering Undersegmented Ancient Scripts Using Phonetic Prior}, year={2021}, doi={10.1162/tacl_a_00354}, volume={9}, journal={Transactions of the Association for Computational Linguistics}, pages={69--81}, author={Luo, Jiaming and Hartmann, Frederik and Santus, Enrico and Barzilay, Regina and Cao, Yuan} }

<rdf:RDF xmlns:dcterms="" xmlns:dc="" xmlns:rdf="" xmlns:bibo="" xmlns:dspace="" xmlns:foaf="" xmlns:void="" xmlns:xsd="" > <rdf:Description rdf:about=""> <dcterms:title>Deciphering Undersegmented Ancient Scripts Using Phonetic Prior</dcterms:title> <dcterms:hasPart rdf:resource=""/> <dc:rights>terms-of-use</dc:rights> <dc:date rdf:datatype="">2022-04-11T14:42:07Z</dc:date> <dc:creator>Cao, Yuan</dc:creator> <dc:creator>Luo, Jiaming</dc:creator> <dc:contributor>Cao, Yuan</dc:contributor> <dc:creator>Hartmann, Frederik</dc:creator> <bibo:uri rdf:resource=""/> <dc:language>eng</dc:language> <void:sparqlEndpoint rdf:resource="http://localhost/fuseki/dspace/sparql"/> <dc:contributor>Luo, Jiaming</dc:contributor> <dcterms:abstract xml:lang="eng">Most undeciphered lost languages exhibit two characteristics that pose significant decipherment challenges: (1) the scripts are not fully segmented into words; (2) the closest known language is not determined. We propose a decipherment model that handles both of these challenges by building on rich linguistic constraints reflecting consistent patterns in historical sound change. We capture the natural phonological geometry by learning character embeddings based on the International Phonetic Alphabet (IPA). The resulting generative framework jointly models word segmentation and cognate alignment, informed by phonological constraints. We evaluate the model on both deciphered languages (Gothic, Ugaritic) and an undeciphered one (Iberian). The experiments show that incorporating phonetic geometry leads to clear and consistent gains. Additionally, we propose a measure for language closeness which correctly identifies related languages for Gothic and Ugaritic. For Iberian, the method does not show strong evidence supporting Basque as a related language, concurring with the favored position by the current scholarship.1</dcterms:abstract> <dc:contributor>Santus, Enrico</dc:contributor> <dc:contributor>Barzilay, Regina</dc:contributor> <dcterms:isPartOf rdf:resource=""/> <dc:creator>Santus, Enrico</dc:creator> <dcterms:rights rdf:resource=""/> <dc:contributor>Hartmann, Frederik</dc:contributor> <dcterms:issued>2021</dcterms:issued> <dcterms:available rdf:datatype="">2022-04-11T14:42:07Z</dcterms:available> <dspace:hasBitstream rdf:resource=""/> <dc:creator>Barzilay, Regina</dc:creator> <dspace:isPartOfCollection rdf:resource=""/> <foaf:homepage rdf:resource="http://localhost:8080/jspui"/> </rdf:Description> </rdf:RDF>

Downloads since Apr 11, 2022 (Information about access statistics)

Luo_2-74mf4c553nsp9.pdf 65

This item appears in the following Collection(s)

Search KOPS


My Account