Time series data motif identification method and device

Abstract

The invention discloses a time series data motif identification method and device, and belongs to the field of time series data analysis. The time series data motif identification method comprises the steps that time series data needing to be analyzed are divided into at least two data subsequences, and each data subsequence is converted into a symbol subsequence; random projection is performed on the symbol subsequences, and the times that each projected symbol subsequence and other projected symbol subsequences have same signals are recorded; two data subsequences which are corresponding to the time exceeding a threshold value in the recorded times and between which the distance is smaller than a first preset distance serve as identified standard motifs; clustering is performed on the standard motifs in each preset range to obtain a center data subsequence, and the variance of each preset range is calculated according to the standard motifs in each preset range and the center data subsequence; the threshold value is decreased, the distances between two data subsequences which are corresponding to the time exceeding the decreased threshold value in the recorded times and the center data subsequence in the preset range where the two data subsequences are located, and the data subsequence of which the distance is smaller than the variance of the preset range where the data subsequence is located serves as the identified motif. According to the time series data motif identification method and device, under the condition that the motif identification speed is guaranteed, the motif identification accuracy can be improved.

Description

technical field [0001] The invention relates to the field of time-series data analysis, in particular to a method and device for pattern recognition of time-series data. Background technique [0002] With the development of statistics, more and more data are in the form of time series data. Wherein, time series data refers to data recorded in time order. For example, daily fluctuation data of the stock market, annual rainfall data, annual traffic condition data, etc. In these time series data, there are some recurring similar subsequences, and these recurring similar subsequences are called motifs. Since the motifs in time series data are of great significance to scientific research, how to identify motifs in large-scale time series data is the key to the study of time series data. [0003] In the existing motif recognition methods for time series data, there are two common recognition methods: precise recognition method and probabilistic recognition method. For the prec...

AI Technical Summary

Problems solved by technology

[0006] Since the above-mentioned accurate identification method needs to calculate the Euclidean distance between every two data subsequences scanned, when the number of scanned data subsequences is large, calculate the Euclidean distance between every two scanned data subsequences The calculation of the K-distance is very heavy, resulting in slow recognition of motifs

[0007] However, the above probabilistic recognition method discretizes and reduces the dimensionality of the time series data through symbolization and random projection, which leads to the loss of some time series data information that may become motifs, and causes a large displacement difference and the original time series that needs to be analyzed Data subsequences with a low probability of repeated occurrence in the data may be identified as motifs. Therefore, the accuracy of the motifs identified by the probabilistic identification method is not high; in addition, when a symbol subsequence and other symbol subsequences are projected When the number of times with the same symbol at the position reaches a certain threshold, the corresponding data subsequence is recognized as a motif, resulting in some motifs that have the same symbol with other symbol subsequences at the projected position that do not reach the threshold cannot be recognized. , so the number of motifs identified by the probabilistic identification method is limited, which further reduces the accuracy of motif identification

Images