Feature recognition of the futures time series based on perceptually important points

Academic and industrial circles use a variety of techniques to reduce the time sequence complexity, a commonly used method is to convert the time sequence into another kind of easy to understand the important points method based on visual expression. Meanwhile in order to help overcome the dimension disaster of time series data, a corresponding index structure should be proposed based on the expressing method. To meet the demand of identify the futures time series which have certain characteristics, this paper proposes ZPIP importance point recognition method based on the Perceptually Important Points, then on this basis, put forward FZPIP feature recognition method based on perceptually important points by introducing the definition of futures trend characteristics. During the process of feature recognition, we obtain the candidate time series with characteristics condition. The paper presents FIS index mechanism based on the binary search tree to solve the inefficient issue of recognition process in real-time. Finally, the experimental results demonstrate the good performance of the proposed method on the futures trading data.