We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.CL

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Computation and Language

Title: DiscoTK: Using Discourse Structure for Machine Translation Evaluation

Abstract: We present novel automatic metrics for machine translation evaluation that use discourse structure and convolution kernels to compare the discourse tree of an automatic translation with that of the human reference. We experiment with five transformations and augmentations of a base discourse tree representation based on the rhetorical structure theory, and we combine the kernel scores for each of them into a single score. Finally, we add other metrics from the ASIYA MT evaluation toolkit, and we tune the weights of the combination on actual human judgments. Experiments on the WMT12 and WMT13 metrics shared task datasets show correlation with human judgments that outperforms what the best systems that participated in these years achieved, both at the segment and at the system level.
Comments: machine translation evaluation, machine translation, tree kernels, discourse, convolutional kernels, discourse tree, RST, rhetorical structure theory, ASIYA
Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
MSC classes: 68T50
ACM classes: I.2.7
Journal reference: WMT-2014
Cite as: arXiv:1911.12547 [cs.CL]
  (or arXiv:1911.12547v1 [cs.CL] for this version)

Submission history

From: Preslav Nakov [view email]
[v1] Thu, 28 Nov 2019 06:05:12 GMT (396kb,D)

Link back to: arXiv, form interface, contact.