Relational Markov Networks for Collective Information Extraction
Relational Markov Networks for Collective Information Extraction
复制标题
用于集体信息提取的关系马尔可夫网络
DOI：
--
复制
发表时间：
2004
期刊：
影响因子：
--
通讯作者：
Razvan Bunescu and Raymond J. Mooney
中科院分区：
文献类型：
--
作者：
Razvan Bunescu and Raymond J. Mooney
大多数信息提取（IE）系统将单独的潜在提取视为独立的。然而，在许多情况下，考虑不同潜在提取之间的影响可以提高整体准确性。基于无向图模型的统计方法，如条件随机场（CRF），已被证明是一种有效的方法来学习准确的IE系统。我们提出了一种新的IE方法，采用关系马尔可夫网络，它可以表示提取之间的任意依赖关系。这允许“集体信息提取”，利用可能的提取之间的相互影响。在从生物医学文本中提取蛋白质名称的实验中，证明了该方法的优越性。
Most information extraction (IE) systems treat separate potential extractions as independent. However, in many cases, considering influences between different potential extractions could improve overall accuracy. Statistical methods based on undirected graphical models, such as conditional random fields (CRFs), have been shown to be an effective approach to learning accurate IE systems. We present a new IE method that employs Relational Markov Networks, which can represent arbitrary dependencies between extractions. This allows for “collective information extraction” that exploits the mutual influence between possible extractions. Experiments on learning to extract protein names from biomedical text demonstrate the advantages of this approach.