首页> 外文会议>10th Workshop on statistical machine translation 2015 >Strategy-Based Technology for Estimating MT Quality
【24h】

Strategy-Based Technology for Estimating MT Quality

机译:基于策略的MT质量评估技术

获取原文
获取原文并翻译 | 示例

摘要

This paper introduces our SAU-KERC system that achieved F1 score of 0.39 in the world-level quality estimation task in WMT2015. The goal is to assign each translated word a "OK" or "BAD" label indicating translation quality. We adopt the sequence labeling model, conditional random fields (CRF), to predict the labels. Since "BAD" labels are rare in the training and development sets, recognition rate of "BAD" is low. To solve this problem, we propose two strategies. One is to replace "OK" label with sub-labels to balance label distribution. The other is to reconstruct the training set to include more "BAD" words.
机译:本文介绍了我们的SAU-KERC系统,该系统在WMT2015的世界级质量评估任务中获得了0.39的F1分数。目的是给每个翻译的单词分配一个“ OK”或“ BAD”标签,指示翻译质量。我们采用序列标记模型,条件随机字段(CRF)来预测标记。由于“ BAD”标签在训练和开发集中很少见,因此“ BAD”的识别率很低。为了解决这个问题,我们提出了两种策略。一种是用子标签替换“确定”标签,以平衡标签分配。另一种是重建训练集以包括更多的“ BAD”单词。

著录项

  • 来源
  • 会议地点 Lisbon(PT)
  • 作者单位

    Knowledge Engineering and Human-Computer Interaction center, Shenyang Aerospace University, Shenyang, China;

    Knowledge Engineering and Human-Computer Interaction center, Shenyang Aerospace University, Shenyang, China;

    Knowledge Engineering and Human-Computer Interaction center, Shenyang Aerospace University, Shenyang, China;

  • 会议组织
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类
  • 关键词

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号