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 Concept Videos

Neuron Structure01:31

Neuron Structure

228.4K
Overview
228.4K

You might also read

Related Articles

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

Sort by
Same author

Impact of nanoparticle morphologies on property prediction using explainable AI.

Nanoscale horizons·2025
Same author

Fractal Characterization of Simulated Metal Nanocatalysts in 3D.

Small science·2025
Same author

Materials "Economatics": Combining Chemical, Financial, Environmental, and Social Factors Using Machine Learning.

ACS nano·2025
Same author

Property Prediction for Complex Compounds Using Structure-Free Mendeleev Encoding and Machine Learning.

Journal of chemical information and modeling·2024
Same author

Classification of battery compounds using structure-free Mendeleev encodings.

Journal of cheminformatics·2024
Same author

Complex Dispersion of Detonation Nanodiamond Revealed by Machine Learning Assisted Cryo-TEM and Coarse-Grained Molecular Dynamics Simulations.

ACS nanoscience Au·2023

Related Experiment Video

Updated: Nov 19, 2025

Silicon Metal-oxide-semiconductor Quantum Dots for Single-electron Pumping
14:58

Silicon Metal-oxide-semiconductor Quantum Dots for Single-electron Pumping

Published on: June 3, 2015

15.1K

Unsupervised structure classes vs. supervised property classes of silicon quantum dots using neural networks.

Amanda J Parker1, Amanda S Barnard

  • 1CSIRO Data61, Door 34 Goods Shed Village St, Docklands, Victoria, Australia.

Nanoscale Horizons
|February 2, 2021
PubMed
Summary

Machine learning classification of silicon quantum dots reveals a disconnect between assumed and data-supported classes. Structural properties alone do not reliably predict optical emission color, challenging intuitive assumptions.

More Related Videos

Resonance Fluorescence of an InGaAs Quantum Dot in a Planar Cavity Using Orthogonal Excitation and Detection
12:57

Resonance Fluorescence of an InGaAs Quantum Dot in a Planar Cavity Using Orthogonal Excitation and Detection

Published on: October 13, 2017

9.4K
Production and Targeting of Monovalent Quantum Dots
10:16

Production and Targeting of Monovalent Quantum Dots

Published on: October 23, 2014

25.8K

Related Experiment Videos

Last Updated: Nov 19, 2025

Silicon Metal-oxide-semiconductor Quantum Dots for Single-electron Pumping
14:58

Silicon Metal-oxide-semiconductor Quantum Dots for Single-electron Pumping

Published on: June 3, 2015

15.1K
Resonance Fluorescence of an InGaAs Quantum Dot in a Planar Cavity Using Orthogonal Excitation and Detection
12:57

Resonance Fluorescence of an InGaAs Quantum Dot in a Planar Cavity Using Orthogonal Excitation and Detection

Published on: October 13, 2017

9.4K
Production and Targeting of Monovalent Quantum Dots
10:16

Production and Targeting of Monovalent Quantum Dots

Published on: October 23, 2014

25.8K

Area of Science:

  • Materials Science
  • Quantum Dots
  • Nanotechnology

Background:

  • Machine learning classification aids in predicting structure/property relationships in nanomaterials.
  • Supervised (human intelligence) and unsupervised (artificial intelligence) methods exist for class separation.
  • The reliability and comparison of these approaches are key research questions.

Purpose of the Study:

  • To explore the utility of different class types in machine learning for nanomaterials.
  • To compare domain-driven (supervised) and data-driven (unsupervised) classification approaches.
  • To investigate the relationship between structural characteristics and optical emission in silicon quantum dots.

Main Methods:

  • Utilized an ensemble dataset of electronic structure simulations for hydrogen-passivated silicon quantum dots.
  • Included data on size, shape, and peak wavelength of optical emission.
  • Employed artificial neural networks for classification analysis.

Main Results:

  • A disconnect was observed between visual optical emission (color) and data-supported classifications.
  • Controlling a limited set of structural characteristics was insufficient for classifying quantum dots by color.
  • Unsupervised machine learning approaches revealed classifications not intuitively apparent from optical properties.

Conclusions:

  • Intuitive assumptions linking specific color bands to distinct quantum dot classes are not always supported by data.
  • Predicting silicon quantum dot optical emission solely based on limited structural features is unreliable.
  • Machine learning, particularly unsupervised methods, offers a powerful tool to uncover complex relationships in nanomaterial properties.