从文本中获取植物知识方法的研究

Botanical Knowledge Acquisition from Text

知识获取一直是人工智能中的一个关键问题。当前,知识的文本挖掘(KAT)已经成为计算机领域的一个重要的研究课题。本文中,给出了基于植物本体的从海量网页文本库中自动获取植物领域知识的方法。该方法包括两个部分,一是植物本体(BotanicalOntology),它是顾芳博士等建立的生物本体的扩展。第二部分是以植物本体为基础,在网络文本库中进行文本挖掘(TextMining),自动获取植物知识。实验证明,基于本体的文本挖掘是一种有效的知识获取方法。

Knowledge acquisition is the bottleneck of artificial intelligence. Knowledge acquisition from text （KAT） is one of the most researched topics in the current computer domain. In this paper, we present an Ontology-based method for acquiring the knowledge of botany from a large corpus of Web pages. The method consists of two components. One is an Ontology of botany. This Ontology is an extension of Gu et al. ＇s Ontology, and is an important base for the second component-the knowledge acquisition from Web pages, our knowledge acquisition method is Ontology-driven in the sense that the content in the Ontology is an efficient guide for what knowledge to extract to fill in the frames in the Ontology.