搜索结果: 46-60 共查到“知识库 机器翻译”相关记录157条 . 查询时间(1.92 秒)
Tree-based translation models, which ex-ploit the linguistic syntax of source lan-guage, usually separate decoding into two steps: parsing and translation. Although this separation makes tree-based de...
The ICT Statistical Machine Translation System for IWSLT 2010
ICT Statistical Machine Translation System IWSLT 2010
2013/4/22
This paper illustrates the ICT Statistical Machine Transla-tion system used in the evaluation campaign of the Interna-tional Workshop on Spoken Language Translation 2010. We participate in the DIALOG ...
The past several years have witnessed rapid ad-vances in syntax-based machine translation, which exploits natural language syntax to guide transla-tion. Depending on the type of input, most of these e...
Mr. Smith, please pay 3,000 dollars for the computer.
As tokenization is usually ambiguous for many natural languages such as Chinese and Korean, tokenization errors might po-tentially introduce translation mistakes for translation systems that rely on 1...
Dependency Forest for Statistical Machine Translation
Dependency Forest Statistical Machine Translation
2013/4/22
We propose a structure calleddependency forest for statistical machine translation.A dependency forest compactly represents multiple dependency trees. We develop new algorithms for extracting string-t...
Dependency-Based Bracketing Transduction Grammar for Statistical Machine Translation
Dependency-Based Bracketing Transduction Grammar Statistical Machine Translation
2013/4/22
In this paper, we propose a nove dependency-based bracketing transduc-tion grammar for statistical machine translation, which converts a source sen-tence into a target dependency tree. Dif-ferent from...
An Efficient Shift-Reduce Decoding Algorithm for Phrased-Based Machine Translation
Efficient Shift-Reduce Decoding Algorithm Phrased-Based Machine Translation
2013/4/22
In statistical machine translation, decod-ing without any reordering constraint is an NP-hard problem. Inversion Transduc-tion Grammars (ITGs) exploit linguistic structure and can well balance the nee...
Effective Constituent Projection across Languages
Effective Constituent Projection across Languages
2013/4/22
We describe an effective constituent pro-jection strategy, where constituent pro-jection is performed on the basis of de-pendency projection. Especially, a novel measurement is proposed to evaluate th...
Traditional 1-best translation pipelines suffer a major drawback: the errors of 1-best outputs, inevitably introduced by each module, will propagate and accumulate along the pipeline. In order to alle...
从双语语料中获取翻译模板
翻译模板 等价对 错误驱动
2010/1/12
翻译模板自动获取是提高MT译文输出质量和领域适应能力的关键性因素。利用Tree-to-String方法抽取等价对，使用错误驱动的学习方法从中获取翻译模板并进行优化。将优化后的翻译模板用于一个基于转换的机器翻译系统中，同时使用“863”对话语料对其进行评测。实验结果表明：当使用自动获取并经优化的模板进行翻译时，开放测试语料的译文评测分数有一定程度的提高。
句法调序的统计机器翻译方法研究
统计机器翻译 句法调序 预先翻译
2010/1/12
为解决基于短语统计机器翻译存在的调序能力不足的问题，尝试利用句法分析器对基于短语统计机器翻译的输入汉语句子进行句法分析，然后利用转换器进行调序操作，并对部分类型短语进行预先翻译，然后再利用基于短语统计机器翻译的解码器进行翻译。重点测试了汉语中“的”字引导的复杂定语调序、介词短语、特定搭配短语、方位词短语的调序及预翻译产生的效果。实验结果表明，这些调序及预翻译操作可以显著地提高基于短语的统计机器翻译...
协同训练支持向量机对遥感影像的分类研究
遥感图像 纹理分析 机器学习
2009/11/19
协同训练可以提高半监督分类器的分类精度，而如何构建具有冗余特性的训练集是其关键所在。依据遥感影像的纹理特征，提出了基于纹理特征值及像素灰度值构建的两个训练集上协同训练支持向量机的算法CTSVMTRS。仿真实验比较了在不同训练集上CTSVMTRS的分类效果，在叠代训练过程中，两类数据集的所有过程的测试结果都存在的明显差异验证了提出的观念。
基于遗传算法的专业元搜索引擎
元搜索引擎 专业搜索引擎 遗传算法
2009/11/17
元搜索引擎返回的查询结果来自独立搜索引擎，要评价此类结果的专业相关性，必须挖掘其位置信息以外的其他信息。研究并实现面向培训领域的元搜索引擎，在充分挖掘网页文本信息的基础上，提取专业网页样本特征，结合遗传算法给出网页专业相关度算法。实验结果表明，该引擎具有较高的专业信息筛选和排序能力。