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

Monte Carlo algorithm for least dependent non-negative mixture decomposition.

Sergey A Astakhov1, Harald Stögbauer, Alexander Kraskov

  • 1John von Neumann Institute for Computing, Forschungszentrum Jülich, D-52425, Jülich, Germany. astakhov@gmail.com

Analytical Chemistry
|March 1, 2006
PubMed
Summary

We developed stochastic non-negative independent component analysis (SNICA), a novel algorithm for separating non-negative sources from mixtures. SNICA outperforms existing methods in analytical spectroscopy and multivariate curve resolution.

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

Electrophysiologic Characteristics of Epidural Spinal Signals in Preclinical Models of Spinal Cord Stimulation.

Neuromodulation : journal of the International Neuromodulation Society·2026
Same author

Electrophysiological Characteristics of Epidural Spinal Signals in Preclinical Models of Spinal Cord Stimulation.

bioRxiv : the preprint server for biology·2026
Same author

From static to dynamic: how object rotation influences grasp decisions in ambiguous settings.

Experimental brain research·2025
Same author

A novel method to assess motor planning deficits in patients with parkinson's disease and mild cognitive impairment.

Scientific reports·2025
Same author

Aftermath epidemics: Percolation on the sites visited by generalized random walks.

Physical review. E·2023
Same author

Many universality classes in an interface model restricted to non-negative heights.

Physical review. E·2023

Area of Science:

  • Signal processing
  • Chemometrics
  • Spectroscopy

Background:

  • Blind source separation is crucial for analyzing complex mixtures.
  • Existing methods struggle with non-negative data, common in analytical spectroscopy.
  • Need for robust algorithms in multivariate curve resolution.

Purpose of the Study:

  • To introduce a novel simulated annealing algorithm, stochastic non-negative independent component analysis (SNICA).
  • To enable blind decomposition of linear mixtures with non-negative sources and coefficients.
  • To address limitations of current methods in specific scientific domains.

Main Methods:

  • Utilized simulated annealing with a Metropolis-type Monte Carlo search.
  • Employed mutual information as a cost function to minimize component dependency.

Related Experiment Videos

  • Enforced non-negativity as a hard constraint during decomposition.
  • Implemented elementary moves including shears and rotations in subspaces.

    • Main Results:

    • SNICA demonstrated superior decomposition performance compared to traditional methods.
    • Outperformed principal component analysis-based methods (MILCA, FastICA, RADICAL).
    • Exceeded the performance of chemometrics techniques (SIMPLISMA, ALS, BTEM).
    • Effective for signals with probability densities peaking at zero.

    Conclusions:

    • SNICA offers a significant advancement in blind source separation for non-negative data.
    • The algorithm is particularly well-suited for analytical spectroscopy and multivariate curve resolution.
    • SNICA provides a more accurate and robust solution for complex mixture decomposition.