N-gram posterior probability confidence measures for statistical machine translation: an empirical study

Adria de Gispert; Graeme Blackwood; Gonzalo Iglesias; William Byrne

首页> 外文期刊>Machine translation >N-gram posterior probability confidence measures for statistical machine translation: an empirical study

【24h】

N-gram posterior probability confidence measures for statistical machine translation: an empirical study

机译：统计机器翻译的N-gram后验概率置信度度量：一项实证研究

获取原文

获取原文并翻译 | 示例

掌桥外文数据库（机构版） >>

开具论文收录证明 >>

文献代查 >>

页面导航

摘要
著录项
相似文献
相关主题

摘要

We report an empirical study of n-gram posterior probability confidence measures for statistical machine translation (SMT). We first describe an efficient and practical algorithm for rapidly computing n-gram posterior probabilities from large translation word lattices. These probabilities are shown to be a good predictor of whether or not the n-gram is found in human reference translations, motivating their use as a confidence measure for SMT. Comprehensive n-gram precision and word coverage measurements are presented for a variety of different language pairs, domains and conditions. We analyze the effect on reference precision of using single or multiple references, and compare the precision of posteriors computed from k-best lists to those computed over the full evidence space of the lattice. We also demonstrate improved confidence by combining multiple lattices in a multi-source translation framework.

机译：我们报告了统计机器翻译（SMT）的n克后验概率置信度测度的经验研究。我们首先描述一种有效而实用的算法，用于从大型翻译词格中快速计算n元语法后验概率。这些概率显示出可以很好地预测是否在人类参考译文中找到了n-gram，从而激发了它们被用作SMT的置信度度量的可能性。针对各种不同的语言对，域和条件，提供了全面的n-gram精度和单词覆盖率测量。我们分析了使用单个或多个引用对引用精度的影响，并将从k最佳列表计算出的后验精度与在整个晶格证据空间上计算出的后代精度进行了比较。通过在多源翻译框架中组合多个晶格，我们还展示了增强的信心。

著录项

来源
《Machine translation》 |2013年第2期|85-114|共30页
作者
Adria de Gispert; Graeme Blackwood; Gonzalo Iglesias; William Byrne;
展开▼
作者单位

Machine Intelligence Laboratory, Department of Engineering, Cambridge University, Cambridge, UK;

IBM T.J. Watson Research, Yorktown Heights, NY 10598, USA;

Machine Intelligence Laboratory, Department of Engineering, Cambridge University, Cambridge, UK;

Machine Intelligence Laboratory, Department of Engineering, Cambridge University, Cambridge, UK;

展开▼
收录信息美国《工程索引》(EI);
原文格式 PDF
正文语种 eng
中图分类
关键词
Statistical machine translation; Minimum Bayes-risk decoding; Confidence measures; N-gram posterior probabilities;

机译：统计机器翻译;最小贝叶斯风险解码;信心措施;N-gram后验概率;

相似文献

外文文献
中文文献
专利

1. N-gram posterior probability confidence measures for statistical machine translation: an empirical study [J] . Adrià Gispert, Graeme Blackwood, Gonzalo Iglesias, Machine Translation . 2013,第2期

机译：统计机器翻译的N-gram后验概率置信度度量：一项实证研究
2. Converting Continuous-Space Language Models into N-gram Language Models with Efficient Bilingual Pruning for Statistical Machine Translation [J] . RUI WANG, MASAO UTIYAMA, ISAO GOTO, ACM transactions on Asian language information processing . 2016,第3期

机译：通过高效的双语修剪将连续空间语言模型转换为N-gram语言模型以进行统计机器翻译
3. Factored bilingual n-gram language models for statistical machine translation [J] . Josep M. Crego, Francois Yvon Machine translation . 2010,第2期

机译：统计机器翻译的因子双语n-gram语言模型
4. Bayes Decision Rules and Confidence Measures for Statistical Machine Translation [C] . Nicola Ueffing, Hermann Ney International Conference on Advances in Natural Language Processing(EsTAL 2004); 20041020-22; Alicante(ES) . 2004

机译：统计机器翻译的贝叶斯决策规则和置信度
5. The relative efficiency of replicates in sweet corn hybrid disease nurseries and posterior probabilities as a measure of confidence in assigning sweet corn hybrids to disease reaction categories. [D] . Michener, Phillip M. 2006

机译：甜玉米杂种疾病苗圃中复制品的相对效率和后验概率，作为对将甜玉米杂种分配给疾病反应类别的信心的度量。
6. Empirical Bayes estimation of posterior probabilities of enrichment: A comparative study of five estimators of the local false discovery rate [O] . Zhenyu Yang, Zuojing Li, David R Bickel 2013

机译：富集后验概率的经验贝叶斯估计：五个本地错误发现率估计量的比较研究
7. N-gram posterior probability confidence measures for statistical machine translation: an empirical study [O] . Adrià de Gispert, Graeme Blackwood, Gonzalo Iglesias, 2012

机译：统计机器翻译的N-gram后验概率置信度度量：一项实证研究

N-gram posterior probability confidence measures for statistical machine translation: an empirical study

摘要

著录项

相似文献

相关主题

期刊订阅