Evaluating the State-of-the-Art Sentence Alignment System on Literary Texts
| Authors | |
|---|---|
| Year of publication | 2021 |
| Type | Article in Proceedings |
| Conference | Recent Advances in Slavonic Natural Language Processing (RASLAN 2021) |
| MU Faculty or unit | |
| Citation | |
| Web | |
| Keywords | Parallel corpora; Automatic alignment; Literary text |
| Description | Sentence alignment is a useful task with many applications in Natural Language Processing and Digital Humanities. This paper presents an evaluation of Vecalign, the state-of-the-art method for automatic sen- tence alignment, on two bilingual corpora built from literary texts. This preliminary study shows that Vecalign performs well for literary texts and gives insights on its remaining issues through a qualitative evaluation of the output alignments. |
| Related projects: |