Paper Info
Title
Finding time series discord based on bit representation clustering
Abstract
The problem of finding time series discord has attracted much attention recently due to its numerous applications and several algorithms have been suggested. However, most of them suffer from high computation cost and cannot satisfy the requirement of real applications. In this paper, we propose a novel discord discovery algorithm BitClusterDiscord which is based on bit representation clustering. Firstly, we use PAA (Piecewise Aggregate Approximation) bit serialization to segment time series, so as to capture the main variation characteristic of time series and avoid the influence of noise. Secondly, we present an improved K-Medoids clustering algorithm to merge several patterns with similar variation behaviors into a common cluster. Finally, based on bit representation clustering, we design two pruning strategies and propose an effective algorithm for time series discord discovery. Extensive experiments have demonstrated that the proposed approach can not only effectively find discord of time series, but also greatly improve the computational efficiency.
Year
DOI
Venue
2013
10.1016/j.knosys.2013.09.015
Knowl.-Based Syst.
Keywords
Field
DocType
effective algorithm,time series discord,bit serialization,bit representation clustering,novel discord discovery algorithm,similar variation behavior,time series discord discovery,main variation characteristic,segment time series,time series,pruning,clustering,dimensionality reduction
Data mining,Dimensionality reduction,Time series data mining,Serialization,Computer science,Theoretical computer science,Artificial intelligence,Merge (version control),Cluster analysis,Machine learning,Piecewise,Computation
Journal
Volume
Issue
ISSN
54
C
0950-7051
Citations 
PageRank 
References 
14
0.65
17
Authors
5
Name
Order
Citations
PageRank
Guiling Li1618.40
Olli Bräysy2137558.39
Liangxiao Jiang381563.17
Zongda Wu425116.20
Yuanzhen Wang58611.78