• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

J4 ›› 2012, Vol. 34 ›› Issue (3): 192-195.

• 论文 • 上一篇    下一篇

基于检索决策树的螨种分类鉴定专家系统

郭冬梅,李朝品   

  1. (安徽理工大学计算机科学与工程学院,安徽 淮南 232001)
  • 收稿日期:2010-11-01 修回日期:2011-04-25 出版日期:2012-03-26 发布日期:2012-03-25
  • 基金资助:

    安徽省自然科学基金资助项目(03043303);安徽省教育厅自然科学基金重点资助项目(2006KJ101A)

An Expert System for Acaridida Classification and Indentification Based on Decision Trees

GUO Dongmei,LI Chaopin   

  1. (School of Computer Science and Engineering,Anhui University of
    Science and Technology,Huainan 232001,China)
  • Received:2010-11-01 Revised:2011-04-25 Online:2012-03-26 Published:2012-03-25

摘要:

应用传统的分类检索方法进行粉螨亚目螨种的分类鉴定是一个复杂、繁琐的过程。随着人工智能技术的发展,采用专家系统技术编制螨种分类鉴定软件将改变传统的分类鉴定方式。通过学习粉螨亚目的分类学和专家系统的相关知识,将螨种的分类鉴定与专家系统有机地结合起来,采用面向对象方法,设计并实现了基于决策树的螨种分类鉴定专家系统。通过构建基于面向对象的决策树知识表示模型,提炼出螨种知识库的记录结构,根据节点对象的逻辑关系,设计了基于决策树的推理机制,并按照一定的推理控制策略实现了螨种的分类鉴定。结果证明,满足了用户的需要,具有重要的实践价值。

关键词: 螨种分类鉴定, 面向对象, 专家系统, 决策树, 推理策略

Abstract:

It is a complicated and tedious process to classify the mites by means of traditional methods. With the development of artificial intelligence,to design an expert system of classification and identification for Acaroid mites will definitely alter the methods of traditional classification.Through studying the relevant knowledge of taxonomy for the species of Acaroid Mites of Acaridida and expert systems, the characterization for mites and expert systems are combined. This paper applies an objectoriented method,designs and implements an expert system of classification and identification for Acaroid mites based on decision trees.According to the knowledge model of objectoriented decision trees, we propose the record strcture of the knowledge base for Acaroid mites, and design a reasoning mechanism based on decision trees by means of the logic of node objects.The reasoning mechanism realizs the identification of the mites by an inferential strategy. The experimental results prove that the system is of practical value, and satisfies the demand of users.

Key words: acaridida classification and identification;objectoriented;expert