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 semi-supervised learning method with hybrid deep belief networks.

Shusen Zhou1, Qingcai Chen2, Xiaolong Wang2

  • 1School of Information and Electrical Engineering, Ludong University, Yantai, Shandong, China.

Plos One
|September 11, 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

The Promotion Effect of Dust on Oxidative Potential of PM<sub>2.5</sub> via Photochemical Aging.

Environmental science & technology·2026
Same author

Crucial Oxidative Stress Risks of Atmospheric Fine Particulate Matter Generated by Sandstorm Process.

Environmental science & technology·2026
Same author

Multi-modal graph neural networks with cross-view attention and contrastive learning for molecular property prediction.

Computer methods in biomechanics and biomedical engineering·2026
Same author

Controllable Protein Design by Prefix-Tuning Protein Language Models.

Journal of chemical information and modeling·2026
Same author

Fourier-Enhanced Kolmogorov-Arnold Network With Attention for Drug-Target Interaction Prediction.

IEEE transactions on computational biology and bioinformatics·2026
Same author

Seasonal dependence of oxidative toxicity of atmospheric fine particulate matter: A case study of Xi'an, a megacity in Northwest China.

Journal of hazardous materials·2026
Same journal

Analysis of strength degradation of coal and rock masses and stability of mined areas under long term immersion environment.

PloS one·2026
Same journal

Biogenic Silver-Selenium nanocomposite with anticancer activity and potent efficacy against vancomycin-resistant Staphylococcus aureus.

PloS one·2026
Same journal

Preparation and physicochemical characterization of a biodegradable chitosan/carboxymethyl cellulose hydrogel synthesized in NaOH/urea medium.

PloS one·2026
Same journal

Action-guilt, survivor-guilt, and depression in combat-related PTSD.

PloS one·2026
Same journal

Explainable machine learning for predicting activities of daily living at discharge in stroke patients: A retrospective study using SHAP interpretability.

PloS one·2026
Same journal

Deep learning based two-way feature depiction model for brain tumor detection.

PloS one·2026
See all related articles

This study introduces active hybrid deep belief networks (AHD), a novel semi-supervised learning method for sentiment classification. AHD effectively abstracts review information using RBM and CRBM layers, outperforming existing algorithms.

Area of Science:

  • Artificial Intelligence
  • Machine Learning
  • Natural Language Processing

Background:

  • Semi-supervised learning is crucial for sentiment classification when labeled data is scarce.
  • Deep learning models offer powerful feature extraction capabilities for text data.

Purpose of the Study:

  • To develop a novel semi-supervised learning algorithm for sentiment classification.
  • To enhance deep learning architectures for effective sentiment analysis using limited labeled data.

Main Methods:

  • A hybrid deep belief network architecture combining Restricted Boltzmann Machines (RBM) and Convolutional Restricted Boltzmann Machines (CRBM).
  • Dimensionality reduction and information abstraction using RBM layers.
  • Effective information abstraction using CRBM layers.

Related Experiment Videos

  • Fine-tuning the deep architecture with gradient descent and an exponential loss function.
  • Integration of active learning to optimize model training with unlabeled data.

    • Main Results:

    • The proposed Active Hybrid Deep Belief Networks (AHD) algorithm demonstrates competitive performance against existing semi-supervised learning methods.
    • Experimental validation on five sentiment classification datasets confirms the algorithm's effectiveness.
    • The study verifies the method's efficacy with varying proportions of labeled and unlabeled reviews.

    Conclusions:

    • AHD offers a robust and effective approach to semi-supervised sentiment classification.
    • The hybrid deep architecture combined with active learning significantly improves sentiment analysis performance.
    • The method is particularly valuable in scenarios with limited labeled data for training.