常识知识查询(英文)  

Querying Commonsense Knowledge

作　　者：胡思康[1] 郑红[2] 

机构地区：[1]北京理工大学计算机科学工程系,北京100081 [2]中国科学院数学与系统科学研究院数学研究所,北京100080

出　　处：《中国科学院研究生院学报》2004年第1期10-18,共9页Journal of the Graduate School of the Chinese Academy of Sciences

基　　金：supportedbytheCASProjectofBrainandMindScience,AgrantfromtheMinistryofScienceandTechnology(2 0 0 1CCA0 3 0 0 0 ),bytheNationalNaturalScienceFoundationofChinaKeyProjectGrand(6973 3 0 2 0 ),InnovationFoundationofAMSS,IOMandICT,andMADISLab

摘　　要：给出了在一个面向Agent、本体和关系的常识知识库中如何实现常识知识查询的过程。在查询过程中,用自然语言形式描述的查询先被转化成一种我们所定义的面向Agent和本体的描述性查询语言。接着,描述性查询被翻译成SQL查询语句。用SQL查询语句对已转化成为关系形式的常识知识库进行查询后,自然语言生成算法把以关系表为形式的查询结果转化成自然语言。实验表明,新的查询方法在查询时间上优于原有的直接在agent和本体上查询知识的旧方法。实验数据表明,查询时间缩短了大约80 %。How to query commonsense knowledge from an Agent, Ontology and Relation oriented Commonsense Knowledge Base (AORKB) is presented. During querying, natural language queries are first transformed into descriptive queries in a kind of language-Agent and Ontology oriented Descriptive Query Language (AODQL) defined by us. Then, descriptive queries are translated into Structured Query Language (SQL) queries. After querying, query results in relation are converted into natural language by the natural language generating module. All tests show that the new method of querying is superior to the method of querying knowledge from agents and ontologies directly in terms of querying time. And up to 80% reduction in querying time is achieved.

关 键 词：面向Agent、本体和关系的常识知识库 面向Agent和本体的描述性查询语言 结构化查询语言 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]