A RECONSTRUCTABLE INTELLIGENT EXPERT SYSTEM FOR TOMATO PLANTING MANAGEMENT IN GREENHOUSE BASED ON CREDIBILITY REASONING
Based on an analysis of the weaknesses in existing agricultural expert systems, a methodology for designing an intelligent expert system based on credibility Neural Network reasoning is introduced. The main components including knowledge acquirement, knowledge presentation, and reasoning are also described. Using several advanced technologies, such as object-oriented, component and so on, the system was designed with features of personalized human-machine information exchange pattern, accurate reasoning and strong learning capabilities. Design patterns and component technology were used in the system and made the framework and architecture of the system more flexible and generalized. Therefore, this system can easily be extended and integrated with other systems. Good effect was gained after it had been put into practice.
Ezpert System Credibility Neural Network Component technology Tomato
Weixing Zhu Morgan M.T. Shuiqing Yu Binjie Zhu Xincheng Li
School of Electrical and Information Engineering, Jiangsu University, Zhenjiang, 212013 Food Science Department, Purdue University, West Lafayette, IN 47907, USA School of Electrial and Information Engineering, Jiangsu University, Zhenjiang, 212013, China Department of Electronic Engineering, Tsinghua University, Beijing, 100084, China School of Mechanical Engineering, Jiangsu University, Zhenjiang,212013, China
国际会议
北京
英文
115-117
2005-10-14（万方平台首次上网日期，不代表论文的发表时间）