SOIL pH FORECAST APPLICATION SYSTEM BASED ON MODIFIED BP NEURAL NETWORK
With continued accretion of the size of agricultural information database, the optimum processing of the information in the database is placed on the order of the day. The paper presents a universal agricultural application- oriented knowledge discovery system. In this system, the genetic algorithm (GA), a general-purpose global search algorithm is used to update the initial weights for minimizing the error between the network output and the desired output. Then based on Levenberg-Marquardt (LM) optimization method, the back- propagation (BP) algorithm is used to further train the neural network prediction model. And then repeat the process until some terminative conditions are sufficed. This method is used to speed up the convergence and to improve the performance. Analyzing and discussing the case of soil pH forecast of Nongan county demonstrated the procedures and performance of this neural network-training algorithm.
Soil pH Back-propagation Genetic algorithm Levenberg-Marquardt
Yang Bai Dayou Liu Chengmin Sun
College of Computer Science and Technology, Jilin University, Changchun 130012, China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education, Jilin University, Changchun 130012, China
国际会议
北京
英文
297-300
2005-10-14（万方平台首次上网日期，不代表论文的发表时间）