一种基于混合模型的时间序列数据挖掘系统

A System of Time Series Data Mining Based on Hybrid Model

时间序列数据挖掘是在从数据仓库中提取出有效历史数据的基础上 ,对其进行一定的处理 ,从而发现隐含、未知的有效信息 .本文阐述了一种新的基于灰色 -回归 -模糊神经网络混合模型的时间序列数据挖掘系统 ,重点讨论了灰色 -回归 -模糊神经网络混合模型的建立过程 ,并应用于浙江省可持续发展预测 ,取得了满意的结果 .该混合模型融合多种智能计算方法优点于一体 ,为时序数据挖掘提供了一种新的实用方法 .

Time series data mining is a process of finding previously hidden, unknown but useful information through extracting and dealing with valid historic records from data warehouse. A new time series data mining system based on Gray Regression Fuzzy Neural Network hybrid model is proposed in this paper and the establishing process of this hybrid model was mainly discussed. This model has been used in forecasting the sustainable development of Zhejiang Province and obtained satisfactory effects. This hybrid model synthesizes the merits of multiple intelligent computation methods and offers a new effective solution of time series data mining.