Validating Meronymy Hypotheses with Support Vector Machines and Graph Kernels
Validating Meronymy Hypotheses with Support Vector Machines and Graph Kernels
复制标题
使用支持向量机和图内核验证 Meronymy 假设
DOI：
--
复制
发表时间：
2010
期刊：
影响因子：
--
通讯作者：
H. Helbig
中科院分区：
文献类型：
--
作者：
Tim vor der Brück;H. Helbig
从文本中提取关系有大量的工作，其中大部分是基于模式匹配或应用树核函数的句法结构。尽管模式应用通常更有效，但当通过F度量进行评估时，树核可以是上级。本文提出了一种基于模式和核函数的部分关系抽取方法。在第一步中，通过应用模式从文本语料库中提取部分命名关系假设。第二步，通过使用几个浅层特征和图核方法来验证这些关系假设。与其他基于表面或句法表示的部分名称提取和验证方法相比，我们使用基于语义网络的纯语义方法。这涉及通过深度句法语义分析器分析维基百科语料库的每个句子，并将其转换为语义网络。通过自动定理证明器从语义网络中提取部分关系假设，该自动定理证明器采用语义网络形式的一组逻辑公理和模式。部分命名候选人，然后通过一个图形内核的方法，基于共同的步行验证。评估表明，该方法实现了相当高的准确率，召回率，和F-措施比使用纯粹的浅层验证的方法。
There is a substantial body of work on the extraction of relations from texts, most of which is based on pattern matching or on applying tree kernel functions to syntactic structures. Whereas pattern application is usually more efficient, tree kernels can be superior when assessed by the F-measure. In this paper, we introduce a hybrid approach to extracting meronymy relations, which is based on both patterns and kernel functions. In a first step, meronymy relation hypotheses are extracted from a text corpus by applying patterns. In a second step these relation hypotheses are validated by using several shallow features and a graph kernel approach. In contrast to other meronymy extraction and validation methods which are based on surface or syntactic representations we use a purely semantic approach based on semantic networks. This involves analyzing each sentence of the Wikipedia corpus by a deep syntactico-semantic parser and converting it into a semantic network. Meronymy relation hypotheses are extracted from the semantic networks by means of an automated theorem prover, which employs a set of logical axioms and patterns in the form of semantic networks. The meronymy candidates are then validated by means of a graph kernel approach based on common walks. The evaluation shows that this method achieves considerably higher accuracy, recall, and F-measure than a method using purely shallow validation.