一种基于Web的通用本体学习框架

提出一种通用的多策略本体学习框架，通过对Web上各专业领域文档集进行挖掘来实现本体自动构建。讨论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，通过实验对算法进行测试和评价。由于集成了多种机器学习算法，该方法在概念抽取和语义关系学习方面具有更高的准确性，采用通用本体WordNet和HowNet作为语料库，可适用于不同的专业领域。通过按需获取Web文档，该方法能实时生成本体。

Web-based General Ontology Learning Framework

This paper proposes a General Ontology Learning Framework(GOLF), dicusses the key technologies of ontology learning such as domain concepts extraction, semantic relationships between concepts and taxonomy automatic construction and ontology evaluation methods. By integrating many machine learning algorithms, this approach suffers less ambiguity and can identify domain concepts and relations more accurately. By using generalized corpus WordNet and HowNet, the method is applicable across different domains. By obtaining source documents from the Web on demand, the method can produce up-to-date ontologies.