Batch-mode semi-supervised active learning for statistical machine translation

doi:10.1016/j.csl.2011.10.001

Batch-mode semi-supervised active learning for statistical machine translation

2013-02-01

21

PRE

AI

S

Sankaranarayanan Ananthakrishnan *

DOI：10.1016/j.csl.2011.10.001

原文链接

原文求助

摘要

En 中文

The development of high-performance statistical machine translation (SMT) systems is contingent on the availability of substantial, in-domain parallel training corpora. The latter, however, are expensive to produce due to the labor-intensive nature of manual translation. We propose to alleviate this problem with a novel, semi-supervised, batch-mode active learning strategy that attempts to maximize in-domain coverage by selecting sentences, which represent a balance between domain match, translation difficulty, and batch diversity. Simulation experiments on an English-to-Pashto translation task show that the proposed strategy not only outperforms the random selection baseline, but also traditional active selection techniques based on dissimilarity to existing training data. (C) 2011 Elsevier Ltd. All rights reserved.

Keyword:

Statistical machine translation

Active learning

Semi-supervised learning

Resource-poor language pairs

Batch-mode semi-supervised active learning for statistical machine translation

摘要

期刊

机构

引用论文