基于小样本时间序列的数据挖掘技术研究

时间序列数据是一类典型的关系型数据，尤其是小样本时间序列数据。针对其样本少、部分信息未知的特点，提出将灰色系统和神经网络相融合，构建灰色神经网络，充分利用两种方法的优势对小样本时间序列数据进行有效挖掘。实验表明：构建的这种网络具有较高的预测精度，非常适用于小样本时间序列数据的挖掘。

Research on Mining Technology for Small Sample Time Series Data

Time-series data is a kind of typical relational data, especially the small sample time-series data. According to the lim-ited samples and part of unknown information, this paper integrates the gray system with neural network and build gray neural network. It makes full use of the advantages of these two methods to exavate small sample time series data mining effectively. Experiments show that the network has higher prediction accuracy. It is quite fit for small sample time-series data mining.