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

COMPOSE: A semisupervised learning framework for initially labeled nonstationary streaming data.

Karl B Dyer, Robert Capo, Robi Polikar

    IEEE Transactions on Neural Networks and Learning Systems
    |May 9, 2014
    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

    iSeqSearch: incremental protein search for iBlast/iMMSeqs2/iDiamond.

    PeerJ·2025
    Same author

    The Naïve Bayes classifier++ for metagenomic taxonomic classification-query evaluation.

    Bioinformatics (Oxford, England)·2024
    Same author

    Complet+: a computationally scalable method to improve completeness of large-scale protein sequence clustering.

    PeerJ·2023
    Same author

    Machine Learning Analysis of Digital Clock Drawing Test Performance for Differential Classification of Mild Cognitive Impairment Subtypes Versus Alzheimer's Disease.

    Journal of the International Neuropsychological Society : JINS·2020
    Same author

    Extensions to Online Feature Selection Using Bagging and Boosting.

    IEEE transactions on neural networks and learning systems·2017
    Same author

    A Sequential Learning Approach for Scaling Up Filter-Based Feature Subset Selection.

    IEEE transactions on neural networks and learning systems·2017
    Same journal

    Granular Ball-Based Noise-Resistant Fuzzy Multineighborhood Feature Selection via Label Enhancement and Feature Graph.

    IEEE transactions on neural networks and learning systems·2026
    Same journal

    Fighting Evolving Spam With ARTMAP Models: A Noise-Resilient Online Detection Framework.

    IEEE transactions on neural networks and learning systems·2026
    Same journal

    HyperSAT: Unsupervised Hypergraph Neural Networks for Weighted MaxSAT Problems.

    IEEE transactions on neural networks and learning systems·2026
    Same journal

    Negation of Basic Belief Assignment in Multisource Information Fusion on Dempster-Shafer Theory With Applications in Pattern Classification.

    IEEE transactions on neural networks and learning systems·2026
    Same journal

    Intervention Feasible Region and Driver Risk Capacity Aware Human-Machine Collaborative Safe Trajectory Planning.

    IEEE transactions on neural networks and learning systems·2026
    Same journal

    A Unified Differential Denoising Learning Framework With a Pre-Trained Model and Fuzzy Graph Networks for Drug-Drug Interaction Prediction.

    IEEE transactions on neural networks and learning systems·2026
    See all related articles

    Compacted Object Sample Extraction (COMPOSE) learns from nonstationary streaming data without labels. This computational geometry framework adapts to changing data distributions, outperforming other methods on synthetic and real-world datasets.

    Area of Science:

    • Machine Learning
    • Data Science
    • Computational Geometry

    Background:

    • Real-world applications increasingly involve streaming data from nonstationary distributions that change over time.
    • Adapting to these changes, known as concept drift, requires new algorithms.
    • Current methods often need extensive labeled data, which is difficult to obtain.

    Purpose of the Study:

    • Introduce Compacted Object Sample Extraction (COMPOSE), a novel framework for learning from nonstationary streaming data.
    • Address scenarios where labels are unavailable or sporadically provided after initialization.
    • Demonstrate COMPOSE's ability to learn and adapt in dynamic data environments.

    Main Methods:

    • Develop a computational geometry-based framework for processing streaming data.

    Related Experiment Videos

  • Implement the COMPOSE algorithm for unsupervised or semi-supervised learning.
  • Evaluate performance on synthetic datasets against optimal Bayes classifiers and arbitrary subpopulation trackers.

    • Main Results:

    • COMPOSE effectively learns from nonstationary streaming data with limited or no labels.
    • The algorithm demonstrates competitive performance against fully supervised methods on real-world datasets.
    • Performance is validated on synthetic datasets simulating various concept drift scenarios.

    Conclusions:

    • COMPOSE offers a robust solution for learning from nonstationary streaming data, particularly when labels are scarce.
    • The framework's adaptability makes it suitable for dynamic, real-world applications.
    • This approach advances unsupervised and semi-supervised learning in the presence of concept drift.