江汉大学学报(自然科学版) ›› 2019, Vol. 47 ›› Issue (4): 378-384.doi: 10.16389/j.cnki.cn42-1737/n.2019.04.015

• 生命科学 • 上一篇    

农作物病虫害专家系统中的知识表示方法

张嫚嫚,张武*,金秀,朱诚,宋一帆,洪汛,李孟杰   

  1. 安徽农业大学 信息与计算机学院,安徽 合肥 230036
  • 出版日期:2019-08-28 发布日期:2019-07-10
  • 通讯作者: 张武
  • 作者简介:张嫚嫚(1993—),女,硕士生,研究方向:人工智能。
  • 基金资助:
    2018 年安徽省重点研究和开发计划项目(1804a07020108);2016 年农业部农业物联网技术集成与应用重点实验室开放基金资助项目(2016KL05);2017 年安徽省重大科技专项计划(17030701049);2019 年安徽省重点研发计划面上攻关项目(201904a0620056)

Knowledge Representation Methods of Expert Systems for Crop Pests and Diseases

ZHANG Manman,ZHANG Wu*,JIN Xiu,ZHU Cheng,SONG Yifan,HONG Xun,LI Mengjie   

  1. School of Information and Computer,Anhui Agriculture University,Hefei 230036,Anhui,China
  • Online:2019-08-28 Published:2019-07-10
  • Contact: ZHANG Wu

摘要: 农作物病虫害的知识表示和知识推理方法的有效性是实现农作物病虫害专家系统正确决策的基础。总结了农作物病虫害专家系统中常见的知识传统表示、知识的现代表示以及知识混合表示方法的优缺点及应用情况,对其发展趋势进行了分析和展望。知识混合表示方法因能有效克服单一表示法的局限性并可充分发挥各种方法的长处而被广泛应用。神经网络知识表示因其快速、准确性高、灵活性高的特点应用前景良好。

关键词: 农作物, 病虫害, 知识表达, 专家系统

Abstract: The effectiveness of knowledge representation methods and knowledge reasoning methods are the basis for the correct decision-making of expert systems for crop pests and diseases. The advantages and disadvantages of traditional representation methods,modern representation methods and hybrid representation methods for common knowledge of expert systems for crop pests and diseases were summarized,and its applications were concluded. Furthermore,the trend of development was analyzed and prospected. The hybrid representation methods applies widely,since they can overcome the limitations of single representation and make full use of the advantages of various methods effectively. The neural network knowledge representation indicates good application prospects due to its fast,high accuracy and high flexibility.

Key words: crops, pests and diseases, knowledge representation, expert system

中图分类号: