自然语言检索的新发展:与Ontology相结合

New Development of Natural Language Information Retrieval: Combination with Ontology

分析了检索语言的发展历程,得出当前流行的检索方式是自然语言检索,但是用自然语言进行检索存在许多弊端,需要情报检索语言加以控制,如何将自然语言的简便性与受控语言的规范性相结合是我们应该深入思考的问题。而本体具有与情报检索语言相类似的一些功能特征,因此提出将Ontology引入自然语言检索系统。本文从4个方面分析了Ontology与自然语言检索结合的必要性,最后利用当前本体语言的推荐标准OWL语言提出了具体的实现方法,得出在自然语言检索的规范控制阶段利用Ontology是提高检索效率的可行方法。

This article analyzes the development process of retrieval language, educing that the popular information retrieval method at present is Natural Language Information Retrieval （NLIR）. However, it has many deficiencies and needs to be controlled by the use of Information Retrieval Language. How to combine the simpleness of natural language with the standardization of controlled language is a problem we should consider earnestly. The Ontology has some similar functions of Information Retrieval Language. Therefore it＇ s proposed to introduce the Ontology into NLIR systems. This article analyzes the necessity of combining the Ontology with NLIR from 4 aspects. Finally, it describes how to realize this combination by the use of OWL language. It concludes that the use of the Ontology to control NLIR is a feasible way to increase the retrieval efficiency.