一种基于组合模型的时间序列数据挖掘系统

提出了一种新的时间序列数据挖掘方法———基于BP(Back Propagation)神经网络和回归分析的组合时间序列数据挖掘模型.重点讨论了神经网络———回归———线性神经网络组合模型的建立过程,强调了通用性,并应用于浙江省可持续发展预测,取得了满意的结果.该组合模型采用神经网络技术来确定权重系数,提高了对复杂非线性系统的拟合能力,为时间序列数据挖掘提供了一种新的实用方法.

A system of time serials data mining based on combined model

A new method of time series data mining,i.e.combined time series data mining model based on BP(Back Propagation) neural network and regression,is proposed in this paper and the establishing process of this combined model is mainly discussed.The general character is also taken into consideration.We apply this model to the forecasting of sustainable development of Zhejiang Province and obtain satisfactory effects.This combined model adopts the technique of neural network to fix weight coefficients,which gives better imitation of complex nonlinear system and offers a new effective solution of time serials data mining.