Time series symbol aggregation approximate representation method fusing trend features

Abstract

The invention discloses a time series symbol aggregation approximate representation method fusing trend features. The time series approximate representation method fusing trend features comprises thefollowing steps: acquiring time series data; preprocessing the time series data; performing time sequence feature segmentation; performing time sequence statistical feature extraction and symbolic representation; performing trend feature extraction and symbolic representation of the time sequence; fusing time series symbol representation and similarity measurement of trend features. According to the method, the trend characteristic information and the statistical characteristic information of a time sequence are combined to form a new symbol aggregation approximate representation method capable of considering the statistical characteristics and the trend characteristics of the time sequence, and the time sequence is mapped from a high-dimensional space to a low-dimensional space on the premise of not losing the sequence characteristic information. Compared with a traditional time sequence representation method, the method not only has better lower bound sealing performance, but also can obtain better classification and clustering effects, thereby better representing time sequences with different morphological characteristics.

Description

technical field [0001] The invention discloses a time series symbol aggregation approximate representation method which combines trend features, and relates to the field of time series data mining. Background technique [0002] Time series data (Time Series, TS) is a common multi-dimensional complex type of data, which objectively records the important information at each observation time point that the observation system changes with time sequence. Time series data has the characteristics of massiveness, high dimensionality, and complexity (noise, unstructured, time axis expansion, linear drift, and discontinuity points), and implies some specific laws and potential characteristics of the observation system. As one of the top ten challenging research problems in the field of data mining, time series data mining is attracting more and more attention from researchers at home and abroad, and is widely used in data mining tasks such as similarity search, pattern mining, and cyc...

AI Technical Summary

Problems solved by technology

[0004] SAX has the advantages of fast dimension reduction and high-efficiency query, but it is also easy to cause the loss of trend feature information inside the original sequence

Especially in the case of relatively large data compression, the larger the data point represented by the sequence mean, the weaker its ability to describe the data sequence information, so it cannot effectively characterize the sequence

Images