ARL Arabic Dependency Treebank

<h3>Introduction</h3><br>
<p>ARL Arabic Dependency Treebank was developed by the US Army Research Laboratory (<a href="https://www.arl.army.mil/www/default.cfm">ARL</a>) and was derived from four LDC resources: Arabic Treebank (ATB) Part 1 v 4.1 (<a href="../../../LDC2010T13">LDC2010T13</a>), Part 2 v 3.1 (<a href="../../../LDC2011T09">LDC2011T09</a>), Part 3 v 3.2 (<a href="../../../LDC2010T08">LDC2010T08</a>) and Broadcast News v 1.0 (<a href="../../../LDC2012T07">LDC2012T07</a>).</p><br>
<p>LDC's ATB series follows the constituency or phrase structure approach to treebank development in which clauses are divided into noun phrases and verb phrases and in each sentence, one or more nodes may correspond to one element. Dependency grammar, on the other hand, is based on the idea that the verb is the center of the clause structure and that other units in the sentence are connected to the verb as directed links or dependencies. This is a one-to-one correspondence: for every element in the sentence there is one node in the sentence structure that corresponds to that element. ARL Arabic Dependency Treebank was generated using constituency-to-dependency software written at ARL.</p><br>
<h3>Data</h3><br>
<p>The source data in this release consists of Arabic newswire and broadcast programming collected by LDC from various news and broadcast providers.</p><br>
<p>The files are in an 11-column tab-separated format with one or more blank lines between sentences. All files are UTF-8 encoded. Further information about the corpus structure is contained in the documentation accompanying this release.</p><br>
<h3>Samples</h3><br>
<p>Please view this <a href="desc/addenda/LDC2016T18.txt">sample</a>.</p><br>
<h3>Updates</h3><br>
<p>None at this time.</p></br>
Portions © 2007-2008 Abu Dhabi TV, © 2000 Agence France Presse, © 2007 Al Alam News Channel, © 2007-2008 Al Arabiya, © 2008 Al Baghdadya TV, © 2008 Al Fayha, © 2007-2008 Al Iraqiyah, © 2005-2007 Aljazeera, © 2007 Al Ordiniyah, © 2008 Al Sharqiyah, © 2002 An Nahar, © 2005, 2007-2008 Dubai TV, © 2007 Kuwait TV, © 2008 Oman TV, © 2007 PAC Ltd, © 2007-2008 Saudi TV, © 2007 Syria TV, © 2001-2002 Ummah Press, © 2003, 2004, 2005, 2007, 2008, 2009, 2010, 2011, 2012, 2016 Trustees of the University of Pennsylvania

<h3>介绍</h3><br>
<p>ARL阿拉伯语依存树库（ARL Arabic Dependency Treebank）由美国陆军研究实验室（<a href="https://www.arl.army.mil/www/default.cfm">ARL</a>）开发，其数据源源自四项语言数据联盟（Linguistic Data Consortium，简称LDC）资源：阿拉伯语树库（Arabic Treebank，简称ATB）第1卷v4.1（<a href="../../../LDC2010T13">LDC2010T13</a>）、第2卷v3.1（<a href="../../../LDC2011T09">LDC2011T09</a>）、第3卷v3.2（<a href="../../../LDC2010T08">LDC2010T08</a>）以及广播新闻语料v1.0（<a href="../../../LDC2012T07">LDC2012T07</a>）。</p><br>
<p>LDC的ATB系列遵循成分句法（又称短语结构）的树库构建范式，该范式将分句拆解为名词短语与动词短语，且单句中可存在一个或多个节点对应同一语言元素。而依存句法则以动词为分句结构的核心，句中其余单元通过有向链接（依存关系）与动词相连，二者为一一对应关系：句子中的每个语言元素均对应句法语结构中的唯一节点。ARL阿拉伯语依存树库由ARL自研的成分转依存句法转换软件生成。</p><br>
<h3>数据</h3><br>
<p>本次发布的源数据包含LDC从多家新闻与广播机构采集的阿拉伯语新闻专线稿件与广播节目内容。</p><br>
<p>所有文件采用11列制表符分隔格式，句子间以一行或多行空行分隔，且均以UTF-8编码。本语料库的结构细节详见本次发布附带的文档说明。</p><br>
<h3>示例</h3><br>
<p>请查看此<a href="desc/addenda/LDC2016T18.txt">示例文件</a>。</p><br>
<h3>更新说明</h3><br>
<p>暂无相关更新。</p><br>
<p>本语料库部分内容 © 2007-2008 阿布扎比电视台、© 2000 法新社、© 2007 阿拉姆新闻频道、© 2007-2008 阿拉伯电视台、© 2008 巴格达迪亚电视台、© 2008 法哈电视台、© 2007-2008 伊拉克电视台、© 2005-2007 半岛电视台、© 2007 奥尔迪尼亚电视台、© 2008 沙尔基亚电视台、© 2002 《纳赫尔报》、© 2005、2007-2008 迪拜电视台、© 2007 科威特电视台、© 2008 阿曼电视台、© 2007 PAC有限公司、© 2007-2008 沙特阿拉伯电视台、© 2007 叙利亚电视台、© 2001-2002 乌玛通讯社、© 2003、2004、2005、2007、2008、2009、2010、2011、2012、2016 宾夕法尼亚大学董事会。</p>

搜集汇总

背景与挑战

背景概述

ARL阿拉伯语依存树库是由美国陆军研究实验室基于LDC阿拉伯语树库资源开发的依存语法树库，数据来源于阿拉伯语新闻专线和广播新闻，采用11列制表符分隔格式。该数据集适用于词性标注、机器翻译、信息检测和解析等自然语言处理任务，其依存语法方法强调动词在句子结构中的中心地位，与短语结构方法形成对比。