UNDERTOW: multi-level segmentation of real-valued time series

Authors:
Tom Armstrong;Tim Oates
Affiliations:
Department of Computer Science and Electrical Engineering, University of Maryland Baltimore County, Baltimore, Maryland;Department of Computer Science and Electrical Engineering, University of Maryland Baltimore County, Baltimore, Maryland
Venue:
AAAI'07 Proceedings of the 22nd national conference on Artificial intelligence - Volume 2
Year:
2007

Citing 5
Cited 0

Normalized Cuts and Image Segmentation

IEEE Transactions on Pattern Analysis and Machine Intelligence
On the approximation of curves by line segments using dynamic programming

Communications of the ACM
Unsupervised Segmentation of Categorical Time Series into Episodes

ICDM '02 Proceedings of the 2002 IEEE International Conference on Data Mining
A symbolic representation of time series, with implications for streaming algorithms

DMKD '03 Proceedings of the 8th ACM SIGMOD workshop on Research issues in data mining and knowledge discovery
Identifying hierarchical structure in sequences: a linear-time algorithm

Journal of Artificial Intelligence Research

Quantified Score

Hi-index	0.00

Visualization

Abstract

The discovery of meaningful change points, finding segments, in both categorical and real-value data time series is a well-studied problem. Prior segmentation algorithms and tasks operate under overly restrictive assumptions (e.g., a priori knowledge of the number of segments, trivial inputs) and in singular domains (e.g., finding common regions in images, speaker change detection). We introduce a domain-independent algorithm, UNDERTOW, which discovers segment boundaries in real-valued time series and constructs hierarchies of segments to form macro segments.