A skip-ngram approach of automatic evaluation for machine translation using SVM regression

Bo Wang; Tiejun Zhao; Muyun Yang; Sheng Li

A skip-ngram approach of automatic evaluation for machine translation using SVM regression

Bo Wang^*
, Tiejun Zhao
, Muyun Yang
, Sheng Li

^*Corresponding author for this work

Faculty of Computing

Harbin Institute of Technology

Research output: Contribution to journal › Article › peer-review

1 Scopus citations

Abstract

This paper proposes an automatic evaluation metric named SNR for the machine translation system. The metric extends the idea of skip-bigram with larger span and multiple statistics. SVM regression method is also introduced into the metric to tune the weights of statistics. The experimental results show that SNR correlates with human assessments better than mainstream metrics on sentence level. 1548-7741/

Original language	English
Pages (from-to)	1279-1286
Number of pages	8
Journal	Journal of Information and Computational Science
Volume	6
Issue number	3
State	Published - Jun 2009

Keywords

MT Evaluation
SVM Regression
Skip-ngram

Cite this

@article{a80bb252fe81459ca854bb1aaa86de55,

title = "A skip-ngram approach of automatic evaluation for machine translation using SVM regression",

abstract = "This paper proposes an automatic evaluation metric named SNR for the machine translation system. The metric extends the idea of skip-bigram with larger span and multiple statistics. SVM regression method is also introduced into the metric to tune the weights of statistics. The experimental results show that SNR correlates with human assessments better than mainstream metrics on sentence level. 1548-7741/",

keywords = "MT Evaluation, SVM Regression, Skip-ngram",

author = "Bo Wang and Tiejun Zhao and Muyun Yang and Sheng Li",

year = "2009",

month = jun,

language = "英语",

volume = "6",

pages = "1279--1286",

journal = "Journal of Information and Computational Science",

issn = "1548-7741",

publisher = "Binary Information Press",

number = "3",

}

TY - JOUR

T1 - A skip-ngram approach of automatic evaluation for machine translation using SVM regression

AU - Wang, Bo

AU - Zhao, Tiejun

AU - Yang, Muyun

AU - Li, Sheng

PY - 2009/6

Y1 - 2009/6

N2 - This paper proposes an automatic evaluation metric named SNR for the machine translation system. The metric extends the idea of skip-bigram with larger span and multiple statistics. SVM regression method is also introduced into the metric to tune the weights of statistics. The experimental results show that SNR correlates with human assessments better than mainstream metrics on sentence level. 1548-7741/

AB - This paper proposes an automatic evaluation metric named SNR for the machine translation system. The metric extends the idea of skip-bigram with larger span and multiple statistics. SVM regression method is also introduced into the metric to tune the weights of statistics. The experimental results show that SNR correlates with human assessments better than mainstream metrics on sentence level. 1548-7741/

KW - MT Evaluation

KW - SVM Regression

KW - Skip-ngram

UR - https://www.scopus.com/pages/publications/70449090134

M3 - 文章

AN - SCOPUS:70449090134

SN - 1548-7741

VL - 6

SP - 1279

EP - 1286

JO - Journal of Information and Computational Science

JF - Journal of Information and Computational Science

IS - 3

ER -

A skip-ngram approach of automatic evaluation for machine translation using SVM regression

Abstract

Keywords

Other files and links

Fingerprint

Cite this