无指导的开放式中文实体关系抽取.pdfVIP

无指导的开放式中文实体关系抽取* 刘安安，秦兵，刘挺 哈尔滨工业大学计算机科学与技术学院，黑龙江哈尔滨，150001 E-mail: {aaliu, bqin, tliu}@ 摘 要：传统的实体关系抽取需要预先定义关系类型体系，然而定义一个全面的实体关系类型体系是很困 难的。开放式实体关系抽取技术解决了预先定义关系类型体系的问题，但是在中文上的研究还比较少。本 文提出面向大规模网络文本的无指导开放式中文实体关系抽取 （UnCORE ：Unsupervised Chinese Open Entity Relation Extraction for the Web ）方法，首先使用实体之间的距离限制和关系指示词的位置限制获取 候选关系三元组，然后采用全局排序和类型排序的方法来挖掘关系指示词，最后使用关系指示词和句式规 则对关系三元组进行过滤。在获取大量关系三元组的同时，还保证了80% 以上的微观平均准确率。 关键词：开放式实体关系抽取；无指导；关系三元组；关系指示词 Unsupervised Chinese Open Entity Relation Extraction Anan Liu, Bing Qin, Ting Liu School of Computer Science and Technology, Harbin Institute of Technology, Harbin 150001 E-mail: {aaliu, bqin, tliu}@ Abstract: Traditionally, Entity Relation Extraction (RE) methods required a pre-defined set of relation types. But it’s difficult to build a well-defined architecture of the relation types. Open Entity Relation Extraction (ORE) is the task of extracting relation triples from natural language text without pre-defined relation types. This paper presents UnCORE (Unsupervised Chinese Open Entity Relation Extraction for the Web), an unsupervised ORE method which is to discover relation triples from large-scale web text. UnCORE exploits word distance and entity distance constraints to generate candidate relation triples, and then adopts global ranking and domain ranking methods to discover relation words from the candidate relation triples. Finally UnCORE filters candidate relation triples by using the extracted relation words and some sentence rules. Results show that UnCORE extracts large scale relation triples at precision higher than 80%. Keywords: Open Entity Relation Extraction; unsupervised; relation triple; relation word 1 引言