Chinese-English Parallel Sentences Extracted from Patents

<h3>Introduction</h3><br>
<p>Chinese-English Parallel Sentences Extracted from Patents was developed by Chilin (HK) Limited and contains 500,000 sentence pairs of Chinese-English parallel text. This resource is based on the training corpus and test sets developed for the Tokyo-based <a href="http://research.nii.ac.jp/ntcir/index-en.html">NTCIR</a> 2009 &amp; 2010 tasks on <a href="http://ntcir.nii.ac.jp/PatentMTList/">Patent Machine Translation</a>.</p><br>
<h3>Data</h3><br>
<p>The sentences in this release were selected from a larger corpus of than 300,000 Chinese-English parallel patents in different fields according to a number of filtering parameters including word alignment, sentence length and language modeling. They were then automatically segmented and aligned. All text is encoded as UTF-8.</p><br>
<h3>Samples</h3><br>
<p>Please view this <a href="desc/addenda/LDC2016T22.cmn.txt">Chinese sample</a> and <a href="desc/addenda/LDC2016T22.eng.txt">English sample</a>.</p><br>
<h3>Updates</h3><br>
<p>None at this time.</p><br>
<h3>Pricing</h3><br>
<p>Not-for-profit organizations may license this data set for US$25.00 under the LDC Not-for-Profit Membership Agreement or under the LDC User Agreement for Non-Members for use in linguistic research, education and non-commercial technology development. For-profit organizations may license this data for US$5000, discounted to US$4000 for LDC for-profit members, under the Commercial License Agreement for Chinese-English Parallel Sentences Extracted from Patents (LDC2016T22).</p><br>
<p>Current fees in this catalog entry reflect those pertaining to a for-profit organization license. Not-for-profit organizations should contact LDC's&nbsp;<a href="mailto:[e-mail]">Membership Office</a> to license this data set.</p></br>
Portions © 2016 Chilin (HK) Limited, © 2016 Trustees of the University of Pennsylvania

<h3>引言</h3><br><p>《从专利中提取的中英平行句对》（Chinese-English Parallel Sentences Extracted from Patents）由奇麟（香港）有限公司（Chilin (HK) Limited）开发，共包含50万条中英平行文本句对。该数据集依托为总部位于东京的<a href="http://research.nii.ac.jp/ntcir/index-en.html">NTCIR</a> 2009年与2010年<a href="http://ntcir.nii.ac.jp/PatentMTList/">专利机器翻译（Patent Machine Translation）</a>任务开发的训练语料库与测试集构建而成。</p><br><h3>数据</h3><br><p>本次发布的句对源自一个包含30余万条跨领域中英平行专利的大型语料库，经词对齐、句长筛选、语言模型校验等多项过滤参数筛选后得到。随后对筛选出的句对进行自动分词与对齐处理，所有文本均采用UTF-8编码格式。</p><br><h3>示例</h3><br><p>可通过以下链接查看示例：<a href="desc/addenda/LDC2016T22.cmn.txt">中文示例</a>与<a href="desc/addenda/LDC2016T22.eng.txt">英文示例</a>。</p><br><h3>更新记录</h3><br><p>暂无更新记录。</p><br><h3>授权费用</h3><br><p>非营利组织可根据<a href="mailto:[e-mail]">语言数据联盟（Linguistic Data Consortium，简称LDC）</a>非营利会员协议或针对非会员的LDC用户协议，以25美元的许可费用获取本数据集的使用权，仅可用于语言学研究、教育及非商业技术开发。营利组织则需按照《从专利中提取的中英平行句对》（LDC2016T22）商业许可协议获取授权，许可费用为5000美元，若为LDC营利会员则可享受4000美元的优惠价格。</p><br><p>本目录条目所列当前费用为营利组织的授权费用标准。非营利组织如需获取本数据集授权，请联系LDC的<a href="mailto:[e-mail]">会员办公室</a>办理授权手续。</p><br><p>部分内容 © 2016 奇麟（香港）有限公司，© 2016 宾夕法尼亚大学托管委员会</p>

搜集汇总

背景与挑战

背景概述

该数据集包含50万句中文-英文平行句对，专门从专利文档中提取，适用于机器翻译研究。句子基于超过30万份平行专利，通过词对齐、句子长度和语言建模等参数筛选，并经过自动分词和对齐处理，所有文本采用UTF-8编码。