TY - GEN
T1 - Selective sharing for multilingual dependency parsing
AU - Naseem, Tahira
AU - Barzilay, Regina
AU - Globerson, Amir
PY - 2012
Y1 - 2012
N2 - We present a novel algorithm for multilingual dependency parsing that uses annotations from a diverse set of source languages to parse a new unannotated language. Our motivation is to broaden the advantages of multilingual learning to languages that exhibit significant differences from existing resource-rich languages. The algorithm learns which aspects of the source languages are relevant for the target language and ties model parameters accordingly. The model factorizes the process of generating a dependency tree into two steps: selection of syntactic dependents and their ordering. Being largely languageuniversal, the selection component is learned in a supervised fashion from all the training languages. In contrast, the ordering decisions are only influenced by languages with similar properties. We systematically model this cross-lingual sharing using typological features. In our experiments, the model consistently outperforms a state-of-the-art multilingual parser. The largest improvement is achieved on the non Indo-European languages yielding a gain of 14.4%.
AB - We present a novel algorithm for multilingual dependency parsing that uses annotations from a diverse set of source languages to parse a new unannotated language. Our motivation is to broaden the advantages of multilingual learning to languages that exhibit significant differences from existing resource-rich languages. The algorithm learns which aspects of the source languages are relevant for the target language and ties model parameters accordingly. The model factorizes the process of generating a dependency tree into two steps: selection of syntactic dependents and their ordering. Being largely languageuniversal, the selection component is learned in a supervised fashion from all the training languages. In contrast, the ordering decisions are only influenced by languages with similar properties. We systematically model this cross-lingual sharing using typological features. In our experiments, the model consistently outperforms a state-of-the-art multilingual parser. The largest improvement is achieved on the non Indo-European languages yielding a gain of 14.4%.
UR - http://www.scopus.com/inward/record.url?scp=84878184179&partnerID=8YFLogxK
M3 - ???researchoutput.researchoutputtypes.contributiontobookanthology.conference???
AN - SCOPUS:84878184179
SN - 9781937284244
T3 - 50th Annual Meeting of the Association for Computational Linguistics, ACL 2012 - Proceedings of the Conference
SP - 629
EP - 637
BT - 50th Annual Meeting of the Association for Computational Linguistics, ACL 2012 - Proceedings of the Conference
T2 - 50th Annual Meeting of the Association for Computational Linguistics, ACL 2012
Y2 - 8 July 2012 through 14 July 2012
ER -