A method and device for double-layer symbolization of time series

Abstract

The present invention provides a time series double-layer symbolization method and device for air quality data, which can retain the specific time interval that each sub-sequence can last. The method includes: grouping the time series according to the size of the observed values ​​in the time series; determining the size of the symbol set and the range of values ​​corresponding to the symbols in the symbol set through Shannon entropy adaptive clustering for the grouped time series; The feature point sequence of the time series is obtained by describing the length criterion and the gradient slope method; according to the jump relationship of the adjacent feature points of the time series in the value range, the division point of the time series on the time axis is determined; according to the division point of the time series, the time is determined The start and end time values ​​of each subsequence in the sequence and the symbols corresponding to the range of its value range convert the time series into a symbolic sequence including temporal relations. The present invention relates to the field of data processing.

Description

technical field [0001] The invention relates to the field of data processing, in particular to a time series double-layer symbolization method and device for air quality data. Background technique [0002] A time series consists of a set of observations recorded at a specific time, often at regular intervals. However, the time series of continuous numerical values ​​is not conducive to analysis in practical applications. The symbolization of time series is a suitable discretization method for effectively obtaining the internal structure of time series. Time series symbolization is important in engineering and science. It has been widely used in many fields such as sociology and economics. However, in the existing technologies, most of them simply perform clustering or directly specify the size of the symbol set for symbolization, which easily leads to the loss of data information, and it is impossible to feedback the duration of different states of the data, for example: ...

AI Technical Summary

Problems solved by technology

[0006] The technical problem to be solved by the present invention is to provide a time-series double-layer symbolization method and device for air quality data, so as to solve the problem in the prior art that the time-series symbolization process requires artificial setting of parameters and cannot keep each state continuously. question of length of time

Images