3751036
doi
10.5281/zenodo.3751036
oai:zenodo.org:3751036
user-newseye
user-eu
Granroth- Wilding, Mark
University of Helsinki
Pivovarova, Lidia
University of Helsinki
A Comparison of Unsupervised Methods for Ad hoc Cross-Lingual Document Retrieval
Zosa, Elaine
University of Helsinki
info:eu-repo/semantics/openAccess
Creative Commons Attribution 4.0 International
https://creativecommons.org/licenses/by/4.0/legalcode
<p>We address the problem of linking related documents across languages in a multilingual collection. We evaluate three diverse unsupervised methods to represent and compare documents: (1) multilingual topic model; (2) cross-lingual document embeddings; and (3) Wasserstein distance. We test the performance of these methods in retrieving news articles in Swedish that are known to be related to a given Finnish article. The results show that ensembles of the methods outperform the stand-alone methods, suggesting that they capture complementary characteristics of the documents.</p>
Zenodo
2020-05-16
info:eu-repo/semantics/conferencePaper
3751035
user-newseye
user-eu
award_title=NewsEye: A Digital Investigator for Historical Newspapers; award_number=770299; award_identifiers_scheme=url; award_identifiers_identifier=https://cordis.europa.eu/projects/770299; funder_id=00k4n6c32; funder_name=European Commission;
1586938818.897548
131565
md5:f419cfd229d47c8a86a09e76aac3cdc9
https://zenodo.org/records/3751036/files/LREC_cross_lingual_news_linking.pdf
public
10.5281/zenodo.3751035
isVersionOf
doi