Exploiting the Semantic Web for Unsupervised Natural Language Semantic Parsing
Exploiting the Semantic Web for Unsupervised Natural Language Semantic Parsing
复制标题
利用语义网进行无监督自然语言语义解析
DOI：
10.21437/interspeech.2012-84
复制
发表时间：
2012
期刊：
影响因子：
--
通讯作者：
Larry Heck
中科院分区：
文献类型：
--
作者：
Gökhan Tür;Minwoo Jeong;Ye;Dilek Z. Hakkani;Larry Heck
在本文中，我们提出将语义网经验和统计自然语言语义分析模型结合起来。其思想是，通过对结构化网页进行语义解析来填充知识库的过程可以为语言理解任务提供非常有价值的隐式注释。我们挖掘命中这些网页的搜索查询，以便对它们进行语义标注，以构建统计的非监督槽fiLling模型，甚至不需要语义标注指南。我们给出了一些有希望的结果，说明了这个想法可以为电影领域建立一个无监督的时隙fi模型，其中有一些代表性的时隙。此外，对于存在领域内未注释句子的情况，我们还采用了无监督模型自适应。这项工作的另一个关键贡献是使用隐式注释的类似自然语言的查询以完全无监督的方式测试模型的性能。我们认为，这样的方法还可以确保语义解析器和后端知识库之间的语义表示一致。
In this paper, we propose to bring together the semantic web experience and statistical natural language semantic parsing modeling. The idea is that, the process for populating knowledge-bases by semantically parsing structured web pages may pro-vide very valuable implicit annotation for language understanding tasks. We mine search queries hitting to these web pages in order to semantically annotate them for building statistical unsupervised slot filling models, without even a need for a semantic annotation guideline. We present promising results demon-strating this idea for building an unsupervised slot filling model for the movies domain with some representative slots. Fur-thermore, we also employ unsupervised model adaptation for cases when there are some in-domain unannotated sentences available. Another key contribution of this work is using im-plicitly annotated natural-language-like queries for testing the performance of the models, in a totally unsupervised fashion. We believe, such an approach also ensures consistent semantic representation between the semantic parser and the backend knowledge-base.