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

A self-organizing principle for learning nonlinear manifolds.

Dimitris K Agrafiotis1, Huafeng Xu

  • 13-Dimensional Pharmaceuticals, Inc., 665 Stockton Drive, Exton, PA 19341, USA. agrafiotis@3dp.com

Proceedings of the National Academy of Sciences of the United States of America
|November 22, 2002
PubMed
Summary

This study introduces a novel self-organizing algorithm for data embedding. It efficiently preserves data structure in low dimensions, scaling linearly for large datasets.

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

Design of immunogens to present a tumor-specific cryptic epitope.

Scientific reports·2025
Same author

Non-Markovian Dynamic Models Identify Non-Canonical KRAS-VHL Encounter Complex Conformations for Novel PROTAC Design.

JACS Au·2024
Same author

Evaluation of clinical and imaging features for differentiating rhabdomyosarcoma from neuroblastoma in pediatric soft tissue.

Frontiers in oncology·2024
Same author

Performance and Analysis of the Alchemical Transfer Method for Binding-Free-Energy Predictions of Diverse Ligands.

Journal of chemical information and modeling·2023
Same author

Taming multiple binding poses in alchemical binding free energy prediction: the β-cyclodextrin host-guest SAMPL9 blinded challenge.

Physical chemistry chemical physics : PCCP·2023
Same author

Molecular Latent Space Simulators for Distributed and Multimolecular Trajectories.

The journal of physical chemistry. A·2023

Area of Science:

  • Data Science
  • Machine Learning
  • Dimensionality Reduction

Background:

  • Modern scientific research generates vast, high-dimensional datasets with complex relationships.
  • Extracting meaningful structure from these large datasets is a significant challenge.

Purpose of the Study:

  • To develop a self-organizing algorithm for embedding high-dimensional data into a low-dimensional space.
  • To preserve the intrinsic dimensionality and metric structure of the original data.
  • To enable analysis of large-scale datasets intractable by conventional methods.

Main Methods:

  • An iterative pairwise refinement strategy is employed to preserve local geometry.
  • The algorithm maintains minimum separation between distant objects, imposing global structure.

Related Experiment Videos

  • It avoids intensive nearest-neighbor or shortest-path computations.

    • Main Results:

    • The algorithm successfully embeds data into a low-dimensional space, preserving its intrinsic structure.
    • It accurately reproduces geodesic distances without prior estimation.
    • Linear scaling allows application to very large datasets.

    Conclusions:

    • The proposed self-organizing algorithm offers an efficient and scalable solution for data embedding.
    • It effectively reveals the underlying geometry of complex, high-dimensional data.
    • This method advances the analysis of big data in various scientific fields.