Online Prediction Study And Application Of Time Series Data Stream

Recently a new class of data-intensive applications has become widely recognized: applications in which the data is modeled best not as persistent relations but rather as transient data streams. Examples of such applications include financial applications,network monitoring, security, telecommunications data management, web applications, manufacturing, sensor networks, and others. This paper focused on the research for time series data streams.The problem with current sliding window implementation that needs to move data in the sliding window as the window slides, which has poor performance. A re-writable linked window technique is proposed, which adopts a rewriting step to update the data in the window and allows no data movement during the process of updating. The theoretical analysis as well as comparative experiments show, the method earns great promotion in light of efficiency, and can be efficiency real data stream processing. A powerful tool for analyzing nonlinear and non-stationary signal which is called EMD (empirical mode decomposition) method is introduced. Different from the traditional method in doing integral transformation to signal, it decompose signal into several IMFs (intrinsic mode function), which contain and extrude the local characteristic of signal. So the characteristic information of the original signal can be well held by analyzing the IMFs. The experimentation testifies that the EMD method applies to nonlinear and non-stationary signal. At this circumstance, the decomposed IMFs are the frequency components. In fact, EMD is a method which extract characteristic from signal. So in most circumstances, it is combined with other method to analyze signal.The essence of using neural network in prediction is a matter of pattern recognition. The emphasis and difficulty lie in how to extract characteristic of signal efficiently, that is, the problem of seeking classification criteria. The appearance of EMD method provides a new way to solve this problem. Therefore, in this paper combined with empirical mode decomposition and radial basis neural networks establish one online time series data stream prediction model called Online_DSPM is proposed. Through this experimentation it is verified that the method of EMD in conjunction with RBF neural network can not only improve the distinguishing accuracy, but also reduce the learning time of RBF neural network greatly.