利用智能引导和KDML增强可拓模型人机建模能力研究

PDF下载HTML

Title:

Research on enhancing the human-machine modeling ability for an extension model using the intelligent guide and KDML

作者:

毛莉娜¹, 李卫华²

1. 广东轻工职业技术学院 机电工程系, 广东 广州 510300;
2. 广东工业大学 计算机学院, 广东 广州 510006

Author(s):

MAO Li’na¹, LI Weihua²

1. Department of Mechanical and Electrical Engineering, Guangdong Industry Polytechnic, Guangzhou 510300, China;
2. School of Computers, Guangdong University of Technology, Guangzhou 510006, China

关键词:

智能引导; KDML; 自然语言理解; 语义; 可拓模型; Agent; 人机交互; 知网

Keywords:

intelligent guide; KDML; natural language understanding; semantic; extension model; Agent; human-machine interaction; HowNet

分类号:

TP3

DOI:

10.11992/tis.201610017

摘要:

为了利用计算机协助人们建立待求解问题的可拓模型，必须让计算机识别用自然语言描述的问题，而且要理解问题的含义，这是相当困难的任务。本文提出利用人机界面Agent的智能引导并结合知网（HowNet）中的知识系统描述语言（KDML），增强计算机语义处理能力的方法。以求职问题为实践的结果说明了方法的有效性。由于KDML有较强的表示语义信息的作用，通过人机交互也能减轻计算机自然语言理解的困难。因此该方法能将自然语言描述的待求解问题的目标和条件进行分离和形式化，使计算机更有效地建立待求解问题的可拓模型。

Abstract:

When using a computer to establish an extension model for solving problems, the computer must be able to recognize problems described in natural languages, and, in particular, must understand the meaning of the problems. This is a very difficult task. Knowledge database mark-up language (KDML) in HowNet has a strong semantic information expression function. Through human-machine interaction, it can also reduce the difficulty that computers encounter when understanding natural language. A method to enhance the computer’s semantic processing ability, based on the human-machine interface agent’s intelligent guide and KDML, is proposed. The goals and conditions of the problems to be solved described in natural language are separated and formalized, making the computer establish the problem-solving extension model more effectively.