Conference paper Open Access
A Comparison of Unsupervised Methods for Ad hoc Cross-Lingual Document Retrieval
Zosa, Elaine; Granroth- Wilding, Mark; Pivovarova, Lidia
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.
Files
(131.6 kB)
| Name | Size | |
|---|---|---|
|
LREC_cross_lingual_news_linking.pdf
md5:f419cfd229d47c8a86a09e76aac3cdc9 |
131.6 kB | Download |
33
34
views
downloads
| All versions | This version | |
|---|---|---|
| Views | 33 | 33 |
| Downloads | 34 | 34 |
| Data volume | 4.5 MB | 4.5 MB |
| Unique views | 29 | 29 |
| Unique downloads | 31 | 31 |
Indexed in
- Publication date:
- May 16, 2020
- DOI:
-
- Grants:
-
European Commission:
- NewsEye - NewsEye: A Digital Investigator for Historical Newspapers (770299)
- Meeting:
- 12th edition of the Language Resources and Evaluation Conference (LREC 2020)
- Communities:
- License (for files):
- Creative Commons Attribution 4.0 International