摘要
使用本体能克服非基于关键字搜索的查全率查准率低下问题,然而使用部分本体的表达能力或是采用基于布尔检索模型仍然有很大的局限性,特别是面对海量无结构的信息资源时,效果尤为不佳。提出一个语义检索模型SRM,基于成熟的领域本体,对各类资源进行语义标注。把用户的自然语言查询转化为语义描述的检索需求,得到相关的本体知识点,再把用这些知识点标注过的相关的资源一并返回给用户。结果表明,语义检索模型SRM能有效克服非基于关键字搜索的查全率查准率低下问题,是一种研究语义研究的有效方法。
Using ontology and its semantics bring us the capability of overcoming the low recall and low precision of keywords search. However,there are a lot of limitations when only making partial use of full expressive power of ontology-based knowledge representation,or based only on boolean retrieval models,especially in the case of searching a mass of resource on the Web. In this paper we propose a Semantic Retrieval Model (SRM) which is based on a specific domain ontology. It transfers the user's natural language query into formal query,and returns related resource that is annotated with the domain knowledge to the user. The results indicate that the SRM could overcome the low recall and low precision of keywords search and could be an effective measurement for studying semantic retrieval.
出处
《微计算机信息》
2010年第36期258-261,共4页
Control & Automation
关键词
语义检索
本体
映射
语义扩展
semantic retrieval
ontology
mapping
semantic expansion