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

Active learning with imbalanced multiple noisy labeling.

Jing Zhang, Xindong Wu, Victor S Shengs

    IEEE Transactions on Cybernetics
    |August 20, 2014
    PubMed
    Summary
    This summary is machine-generated.

    Related Concept Videos

    Difference from Background: Limit of Detection01:05

    Difference from Background: Limit of Detection

    8.7K
    The limit of detection (LOD) is the smallest amount of analyte that can be distinguished from the background noise. The LOD value corresponds to the concentration at which the analyte signal is three times larger than the standard deviation of the blank signal. Below this value, the analyte signal cannot be differentiated from the background noise. It is calculated by dividing the calibration slope by 3 times the standard deviation of the blank signals.
    The LOD indicates the presence or absence...
    8.7K

    You might also read

    Related Articles

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

    Sort by
    Same author

    Quantitative proteomics identifies surfactant-resistant alpha-synuclein in cerebral cortex of Parkinsonism-dementia complex of Guam but not Alzheimer's disease or progressive supranuclear palsy.

    The American journal of pathology·2007
    Same author

    Different supramolecular assemblies in two 1:1 proton-transfer compounds of sulfobenzoic acids with aromatic amines.

    Acta crystallographica. Section C, Crystal structure communications·2007
    Same author

    Identification of proteins involved in microglial endocytosis of alpha-synuclein.

    Journal of proteome research·2007
    Same author

    Biomarkers for Alzheimer's disease.

    Expert review of neurotherapeutics·2007
    Same author

    [Role of sympathetic nerve activity and arterial endothelial function in pathogenesis of hypertension in patients with obstructive sleep apnea-hypopnea syndrome].

    Zhonghua jie he he hu xi za zhi = Zhonghua jiehe he huxi zazhi = Chinese journal of tuberculosis and respiratory diseases·2007
    Same author

    Subsequently enhanced CPP to morphine following chronic but not acute footshock stress associated with corticosterone mechanism in rats.

    The International journal of neuroscience·2007
    Same journal

    Relaxed Stability Conditions for Model Predictive Control of Hybrid Dynamical Systems Using Hybrid Recurrent Neural Networks.

    IEEE transactions on cybernetics·2026
    Same journal

    An Evolutionary Algorithm Assisted by an Ensemble of Pareto-Optimal Surrogate Models.

    IEEE transactions on cybernetics·2026
    Same journal

    A Quantum Self-Attention Neural Network Model on Quantum Circuits.

    IEEE transactions on cybernetics·2026
    Same journal

    Semi-Explicit Solution of Some Discrete-Time Higher-Order-Cost Mean-Field-Type Control.

    IEEE transactions on cybernetics·2026
    Same journal

    A Novel One-Step Small Object Detector for Autonomous Aerial Vehicles.

    IEEE transactions on cybernetics·2026
    Same journal

    Online Data-Driven-Based Optimal Output Tracking Control Without Initial Stabilizing Policy.

    IEEE transactions on cybernetics·2026
    See all related articles

    This study introduces a new active learning framework to handle multiple noisy labels from crowdsourcing. The proposed methods effectively integrate labels and select instances, improving supervised learning performance even with imbalanced data.

    Area of Science:

    • Machine Learning
    • Computer Science
    • Data Science

    Background:

    • Crowdsourcing systems enable collecting multiple noisy labels for supervised learning.
    • This process aligns with active learning and presents challenges like imbalanced multiple noisy labeling.

    Purpose of the Study:

    • To propose a novel active learning framework for crowdsourcing systems with multiple imperfect annotators.
    • To address the imbalanced multiple noisy labeling problem in supervised learning.

    Main Methods:

    • Introduced a positive label threshold (PLAT) algorithm for label integration, dynamically adjusting thresholds to determine class membership.
    • Developed three novel instance selection strategies: label uncertainty, model uncertainty, and a combined method (CFI).

    Related Experiment Videos

    Main Results:

    • Experimental results on 12 datasets show significant improvements in learning performance with the proposed instance selection strategies.
    • The combined method (CFI) demonstrated the best performance, particularly in scenarios with varying levels of labeling imbalance.
    • The methods achieved high performance when applied to real-world noisy labels from Amazon Mechanical Turk.

    Conclusions:

    • The novel active learning framework effectively handles multiple noisy labels in crowdsourcing.
    • The proposed instance selection strategies, especially CFI, enhance supervised learning performance in imbalanced labeling scenarios.
    • The framework shows practical applicability and high performance in real-world crowdsourcing data.