Incorporation of Linguistic Features in Machine Translation Evaluation of Arabic
This paper describes a study on the contribution of some basic linguistic features to the task of machine translation evaluation of Arabic as a target language. AL-TERp is used as a metric dedicated and tuned especially for Arabic. Performed experiments on a medium sized corpora show that linguistic...
Saved in:
Published in | Big Data, Cloud and Applications pp. 500 - 511 |
---|---|
Main Authors | , , |
Format | Book Chapter |
Language | English |
Published |
Cham
Springer International Publishing
|
Series | Communications in Computer and Information Science |
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | This paper describes a study on the contribution of some basic linguistic features to the task of machine translation evaluation of Arabic as a target language. AL-TERp is used as a metric dedicated and tuned especially for Arabic. Performed experiments on a medium sized corpora show that linguistic knowledge improves the correlation of metric results with human assessments. Also a detailed qualitative analysis of the results highlights a number of resolved issues related to the use of linguistic features. |
---|---|
ISBN: | 3319962914 9783319962917 |
ISSN: | 1865-0929 1865-0937 |
DOI: | 10.1007/978-3-319-96292-4_39 |