Time sequence variable-length die body mining method

Abstract

The invention discloses a time sequence variable-length die body mining method. The method comprises the following steps: 1, die body extraction: using a die body mining algorithm STOMP based on Matrix Profile as a subprogram, introducing a lower bound distance calculation acceleration strategy combined with an incremental distance, and accelerating to find all possible lengths of motifs; 2, die body grouping: adding die body overlapping and length similarity conditions to carry out die body grouping; and 3, die body grouping equivalence class division: adding a die body grouping overlapping condition to carry out equivalence class division on the die body groups; and 4, variable-length die body extraction: extracting motif representatives in each grouping equivalence class, wherein the die body representation set is the variable-length die body. According to the method, too short, too long and commonly matched die bodies can be eliminated, the lengthened die body in the time sequenceis extracted, and the accuracy, efficiency and expandability are improved.

Description

technical field [0001] The invention belongs to the technical field of information processing, and in particular relates to a time series variable length motif mining method. Background technique [0002] Time series motif mining can find recurring similar fragments from time series in an unsupervised manner, find meaningful, novel, and unknown knowledge in the data, and thus discover potential rules and specific events in time series. In addition, time series motif mining is not only applicable to one-dimensional or multi-dimensional data, but also applicable to different types of sequence data, such as spatial sequence data, time series data, and stream data. And time series motif mining technology has also been applied in genetics, medicine, mathematics, music and many other fields. [0003] Motifs are defined as recurring patterns, frequent trends, or near-repeating sequences, shapes, fragments, subsequences, etc. Mueen gave his definition of a motif: a motif is a pair...

AI Technical Summary

Problems solved by technology

In addition, using the MK algorithm as a subroutine to iterate leads to problems such as slow algorithm speed and poor scalability.

Images