摘要
在分析灰色GM(1,1)模型和BP神经网络模型优缺点的基础上,构建了基于GM(1,1)与BP神经网络组合预测模型。首先利用GM(1,1)模型对系统发展进行预测得到一组预测值,同时分别将原始数据与预测数据作为输入输出数据对BP神经网络进行训练,以便得到权值和阀值,最后利用得到的权重和阈值并输入预测年份,即可得预测值。将构建的组合模型对中国人口未来发展趋势进行预测,预测结果表明,人口总量在中短期内继续增长,增速较为平稳,每年以0.11亿人口数增长。该算例表明了该组合预测模型具有较高的预测精度。
The combination forecast model of GM (1, 1) and BP neural network has been constructed based on the analysis of GM (1, 1) model and BP neural network model for the advantages and disadvantages in this paper. Firstly, a set of predicted values is ob-tained by GM(1, 1).Secondly,a index weight was obtained by the original data and forecasting data as input and output ones to train the BP neural network.Finally,prediction can be obtained by using the weights and thresholds. Combination forecast model has been applied to forecast development trend of the population of China in the future,the prediction results show that the population continued to grow in the medium term and relatively stable growth, with 0.11 million population growth each year. The example shows that the combination forecast model has higher prediction accuracy.
出处
《铜陵学院学报》
2016年第3期102-104,共3页
Journal of Tongling University
基金
安徽省高校省级人文社会科学重点研究项目(SK2015537)
安徽省大学生创业训练项目(AH201410383115)
