论文部分内容阅读
为了提高人口预测精度,提出了基于多项式神经网络模型与递推最小二乘法的人口预测方法.方法完全避免了人为假设条件,充分利用我国六次人口普查数据来建立基于多项式神经网络模型的人口预测模型,并使用递推最小二乘算法递推计算多项式神经网络模型的加权系数.方法能有效预测中长期人口数据及其变化趋势.研究结果表明,中国将在2016年达到人口高峰1385亿.
In order to improve the accuracy of population prediction, a population prediction method based on polynomial neural network model and recursive least square method is proposed. The method completely avoids the artificial assumptions and makes full use of the six census data to establish the population prediction based on the polynomial neural network model Model and using the recursive least squares algorithm to recursively calculate the weighted coefficient of the polynomial neural network model.The method can effectively predict the long-term population data and its changing trend.The results show that China will reach a peak of 138.5 billion in 2016.