Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Joint segmentation and classification of time series using class-specific features.

Zhen Jane Wang1, Peter Willett

  • 1Electrical and Computer Department, University of Maryland, College Park, MD 20742, USA. wangzhen@glue.umd.edu

IEEE Transactions on Systems, Man, and Cybernetics. Part B, Cybernetics : a Publication of the IEEE Systems, Man, and Cybernetics Society
|September 21, 2004
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

UCSF RMaC: University of California San Francisco 3D Multi-Phase Renal Mass CT Dataset with Tumor Segmentations.

medRxiv : the preprint server for health sciences·2026
Same author

Factors Associated With Aborted Whipple Procedures for Periampullary Carcinoma: A Multicenter Case-Control Study by the SAR Pancreatic Ductal Adenocarcinoma Disease Focus Panel.

AJR. American journal of roentgenology·2025
Same author

Spatially constrained hyperpolarized 13C MRI pharmacokinetic rate constant map estimation using a digital brain phantom and a U-Net.

Journal of magnetic resonance (San Diego, Calif. : 1997)·2025
Same author

Searching chemical databases in the pre-history of cheminformatics.

Journal of cheminformatics·2024
Same author

Advancements in early detection of pancreatic cancer: the role of artificial intelligence and novel imaging techniques.

Abdominal radiology (New York)·2024
Same author

A Computer Vision Algorithm to Predict Superior Mesenteric Artery Margin Status for Patients With Pancreatic Ductal Adenocarcinoma.

Annals of surgery·2024

This study introduces a novel method for time series segmentation and classification using statistical models. The approach accurately identifies segments without needing to pre-specify their number, offering efficient and precise analysis.

Area of Science:

  • Signal Processing
  • Statistical Modeling
  • Machine Learning

Background:

  • Time series analysis often requires accurate segmentation and classification.
  • Existing methods may necessitate a priori knowledge of segment counts or complex joint statistical characterizations.

Purpose of the Study:

  • To develop a flexible and accurate approach for joint time series segmentation and classification.
  • To enable segmentation without prior specification of the number of segments.

Main Methods:

  • A two-stage approach involving piecewise generalized likelihood ratio (GLR) for initial segmentation and subsequent refinement.
  • Utilizes a menu of statistical models, each defined by a sufficient statistic and its probability density function (PDF).
  • Incorporates penalization to prevent over-segmentation.

Related Experiment Videos

Main Results:

  • The method achieves high accuracy in both segment number discovery and segmentation.
  • Demonstrates a computationally efficient algorithm with a burden approximately linear to the time series length.
  • A hybrid approach with Gibbs sampling offers improved accuracy at a slightly increased computational cost.

Conclusions:

  • The presented method provides an accurate and computationally efficient solution for time series segmentation and classification.
  • Its flexibility in model selection and lack of need for pre-specified segment numbers make it broadly applicable.
  • The hybrid Gibbs sampling extension further enhances accuracy for demanding applications.