Jove
Visualize
Contact Us

Related Experiment Videos

A classification paradigm for distributed vertically partitioned data.

Jayanta Basak1, Ravi Kothari

  • 1IBM India Research Laboratory, Indian Institute of Technology, New Delhi 110016, India. bjayanta@in.ibm.com

Neural Computation
|May 29, 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

Low back pain, with or without sciatica.

Practical neurology·2025
Same author

Multidisciplinary approach to L3/L4 lumbar disc prolapse masquerading as focal limb myositis-a radiological challenge.

BJR case reports·2021
Same author

Assessment of sensitivity of whole body CT for major trauma.

European journal of trauma and emergency surgery : official publication of the European Trauma Society·2018
Same author

Patient attitudes to tonsillectomy.

International journal of family medicine·2013
Same author

Multiple exemplar-based facial image retrieval using independent component analysis.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2006
Same author

Online adaptive decision trees: pattern classification and function approximation.

Neural computation·2006
Same journal

A Model-Free Reinforcement Learning Implementation of Decision Making Under Uncertainty by Sequential Sampling.

Neural computation·2026
Same journal

DROP: Distributional and Regular Optimism and Pessimism for Reinforcement Learning.

Neural computation·2026
Same journal

Hierarchical Active Inference Using Successor Representations.

Neural computation·2026
Same journal

W-Kernel and Its Principal Space for Frequentist Evaluation of Bayesian Estimators.

Neural computation·2026
Same journal

A Hidden Markov Model-Inspired Sequence Classification Method for Hyperdimensional Computing.

Neural computation·2026
Same journal

Sparse Graphical Modeling for Electrophysiological Phase-Based Connectivity Using Circular Statistics.

Neural computation·2026
See all related articles
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

This study introduces a novel classification algorithm for distributed, vertically partitioned data. The method effectively combines local classifiers to minimize error rates, approaching the performance of a centralized classifier.

Area of Science:

  • Computer Science
  • Machine Learning
  • Data Science

Background:

  • Traditional pattern classification assumes complete feature availability.
  • Geographically distributed data often results in servers with partial, non-synchronized feature subsets.
  • Infrastructure limitations hinder centralized data access and synchronization.

Purpose of the Study:

  • To develop a classification algorithm for distributed, vertically partitioned data.
  • To construct a global classifier using pre-existing local classifiers from distributed servers.
  • To minimize the error rate of the global classifier, approaching that of a classifier with full feature access.

Main Methods:

  • Local classifiers are built using partial feature subsets available at each server.

Related Experiment Videos

  • A novel algorithm integrates these local classifiers to form a global classifier.
  • The algorithm aims to optimize performance without requiring data synchronization.

    • Main Results:

    • Empirical validation demonstrates the proposed algorithm's effectiveness.
    • Theoretical results quantify the performance loss compared to a classifier with complete feature access.
    • The algorithm achieves error rates close to an ideal centralized classifier.

    Conclusions:

    • The proposed algorithm offers a viable solution for classification with distributed, vertically partitioned data.
    • It effectively leverages local classifiers to achieve high performance without data centralization.
    • This approach addresses practical challenges in real-world distributed systems.