UNDERTOW: Multi-Level Segmentation of Real-Valued Time Series

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.

Cite

Text

Armstrong and Oates. "UNDERTOW: Multi-Level Segmentation of Real-Valued Time Series." AAAI Conference on Artificial Intelligence, 2007.

Markdown

[Armstrong and Oates. "UNDERTOW: Multi-Level Segmentation of Real-Valued Time Series." AAAI Conference on Artificial Intelligence, 2007.](https://mlanthology.org/aaai/2007/armstrong2007aaai-undertow/)

BibTeX

@inproceedings{armstrong2007aaai-undertow,
  title     = {{UNDERTOW: Multi-Level Segmentation of Real-Valued Time Series}},
  author    = {Armstrong, Tom and Oates, Tim},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year      = {2007},
  pages     = {1842-1843},
  url       = {https://mlanthology.org/aaai/2007/armstrong2007aaai-undertow/}
}