用遗传程序设计预测晚稻普通矮缩病流行趋势

Analyses of Prevalent Trend of Dwarf Disease of Late Rice by Genetic Programming

运用遗传程序设计方法作为建模工具,将全部17个样本(17年的调查数据)分为训练集(15个样本)、预测集(2个样本).研究结果表明,所建模型能完全正确地识别预测样本,即在无任何先验知识的条件下,建立了晚稻普矮病流行趋势预测的计算机专家系统.该遗传程序设计方法只要参数选择适当。

Using the genetic programming approachs as the modelling tool,all 17 samples were divided into a training set (15 samples)and a testing set (2 samples). The result indicated that the prediction for testing set was totally conformed with the actual result by the programme finally obtained. It was indicated that a computer expert system can be established by the genetic programming method without any testified knowledge. As long as the parameters are selected rationally,the genetic programming approach may be expected to be a broader computer method in plant diseases forecasting.