可移植的自然语言生成系统中知识库的设计  

Design of Knowledge Base in Transplantable Natural Language Generation System

作　　者：黄友能[1] 

机构地区：[1]北京交通大学计算机与信息技术学院,北京100044

出　　处：《北京交通大学学报》2004年第5期20-23,共4页JOURNAL OF BEIJING JIAOTONG UNIVERSITY

摘　　要：从可移植的角度对自然语言生成系统中任务域处理进行了研究,结合面向对象的知识库的基本理论,从知识表示、知识获取、知识推理、知识更新等方面提出了可移植的自然语言生成系统中知识库的设计方法,并通过实验结果证明该方法是可行的.In this paper, the transplantable design method of natural language generation system is presented based on the research from task domain processing,a detail knowledge base design, which came from the point of knowledge_representation, knowledge_acquirement, knowledge_consequence, knowledge_refreshment,is proposed in transplantable natural language generation system applied the theory of knowledge base on object. Furthermore, by the experiment, we conclude that the new design method is effective and reasonable, and the preliminary results are provided.

关 键 词：移植性 自然语言 知识库 数据设计 语言生成系统 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]