面向文本的本体学习研究概述

A Survey of Ontology Learning from Text

对本体(ontology)的研究在计算机领域变得越来越广泛,但手工构造本体是一项繁琐而辛苦的任务,还会导致知识获取瓶颈。本体学习技术是利用本体工程技术和机器学习技术等众多学科技术来实现本体的(半)自动构建。本体的学习可以面向文本、知识库、结构化数据、半结构化数据和无结构数据。本文主要介绍了面向文本的本体学习,并对其中的学习内容、学习方法、学习工具、学习过程和系统评价等关键技术进行了说明,特别介绍了学习方法中的基于统计的方法、词汇句法模式法和形式概念分析法并对其优缺点做了简单的分析。

Research on ontology is increasingly becoming widespread in the computer science community. But the manual construction of ontology is a time-consuming task and easily leads to the bottleneck of knowledge acquisition. Ontology learning aims at the integration of a multitude of disciplines such as ontology engineering techniques and machine learning techniques to construct the ontology （semi）automatically. There are different ontology learning approaches according to the type of input： ontology learning from text, from knowledge base, from structured-data, from semistructured data and from unstructured data. Ontology learning from text is mainly introduced. The key technologies of ontology learning from text are presented, including learning content, learning approach, learning tool, learning process and system evaluation. Authors especially introduce the statistical method, lexico-syntactic pattern method and formal concept analysis method and simply analyze the advantage and disadvantage of these methods.