Parameter-free motif discovery for time series data

Abstract

Time series motif discovery is an increasingly popular research area in time series mining whose main objective is to search for interesting patterns or motifs. A motif is a pair of time series subsequences, or two subsequences whose shapes are very similar to each other.Typical motif discovery algorithm requires a predefined motif length as its parameter.Discovering motif with arbitrary lengths introduces another problem, where selecting a suitable length for the motif is nontrivial since domain knowledge is often required. Thus, this work proposes a parameter-free motif discovery algorithm called k -Best Motif Discovery (k BMD) which requires no parameter as input, and as a result returns a set of all "Best Motif " that are ranked by our proposed scoring function which is based on similarity of motif locations and similarity of motif shapes.

Survey on time series motif discovery

“…The k ‐Best Motif Discovery ( k BMD) method applies a sliding window of different lengths several times in order to define motifs of variable lengths without providing the length of the motif. This method performs the MK algorithmand has high time complexity . Lin's grammar‐based method detects motifs by finding the repeated symbolic subsequences in time series with the linear time complexity.…”

“…Detection of motifs of variable length is another problem which is solved by methods such as Lin's grammar‐based method, k BMD, and VLMD . The other methods such as in Refs also employ the same procedure in order to detect motifs of variable length and size: performing sliding windows of different sizes.…”

EXTRACT: Strong Examples from Weakly-Labeled Sensor Data

“…Like most existing work, we find event instances by searching over shorter regions of data within the overall time series (Fig 2a). Since we do not know how long the instances are, this seemingly requires exhaustively searching regions of many lengths [10], [12], [22], so that the instances are sure to be included in the search space.…”

A Review of Subsequence Time Series Clustering

“…A parameter-free motif discovery algorithm called kBMD finds k -best motif in any time series sequence without the need of any parameters. The algorithm returns a small set of motifs, which are ranked by a scoring function [ 73 ].…”

“…Nunthanid et al [ 73 ] propose a parameter-free algorithm for motif discovery called k -best motif discovery (kBMD). This algorithm detects k -best motifs without any parameters.…”