面向数字图书馆的本体学习研究  

A study on Ontology Learning for Digital Library

作　　者：刘柏嵩[1] 

机构地区：[1]宁波大学图书馆,宁波315211

出　　处：《大学图书馆学报》2006年第6期30-34,38,共6页Journal of Academic Libraries

基　　金：浙江省自然科学基金(Y105625);浙江省社科规划项目(N04GL07);宁波市青年博士基金(2005A620005)资助

摘　　要：面向语义Web的智能数字图书馆的实现很大程度上依赖于本体的建立,本体与数字图书馆中的数字资源采集、数字馆藏和用户访问网关都密切相关。在本体构建方面,目前存在的绝大多数本体都是手工生成的,该方法效率低、出错率高,更难以维护和更新。这对语义级数字图书馆的实现造成了巨大的障碍。为此提出了一种面向数字图书馆的本体学习方法GOLF,通过对各专业领域中大量的Web文档集和语料库进行挖掘来实现本体学习,并分别讨论了本体学习中本体概念的抽取、概念之间语义关系的抽取和分类体系的自动构建等关键技术。The Semantic-Oriented digital libraries strongly depend on the construction of ontologies. Digital resources collections and user access gateway of digital libraries are closed related to ontologies. To generate ontologies, an issue named ＂ontology bottleneck＂, which is the lack of efficient ways to build ontologies, has been brought up. Therefore, it is an urgent task to improve the methodology for rapid development of more detailed and specialized domain ontologies. However, it has been a hard task because domain concepts have highly-specialized semantics and the number of concepts is fairly large. In order to reduce the cost, the ontology learning framework GOLF has been developed in our research group. In this paper, we confirm the significance of ontology learning framework. This paper proposes a framework of ontology learning from corpus. Then key technologies of ontology learning such as domain concepts extraction and semantic relationships between concepts and taxonomy automatic construction are discussed.

关 键 词：本体 数字图书馆 本体学习 语义网 

分 类 号：G250.76[文化科学—图书馆学]