摘要
基于粗集、遗传神经网络的环境质量评价方法利用粗集对属性的归约功能将数据库中的数据进行归约,并将归约后的数据作为训练数据提供给BP神经网络;再用遗传算法和BP算法相结合的混合算法来训练网络预测模型的结构(在得到最优网络结构的同时也得到网络的最优权值和阈值)。通过实例表明该方法是有效的,为环境质量评价提供了一种新的研究思路和分析方法。
With the method of environmental quality assess ment based on rough sets and genetic-neural network, the data of the database were reduced by using rough sets reduction function, and then the reduced data were transferred to the BP neural network as training data. The structure of the network prediction model was trained by the genetic algorithm and the BP neural network, to obtain the optimum weights, threshold values and the optimum structure. The results for an example demonstrated that this method is valid for the assessment of environmental quality. This method provides a new concept for the establishment of environmental quality assessment models.
出处
《广州环境科学》
2005年第4期35-39,共5页
Guangzhou Environmental Science
关键词
环境质量评价
粗集
遗传算法
BP神经网络
environmental quality assessment rough sets genetic algorithm BP neural network