2023
DOI: 10.1021/acs.jcim.2c01203
|View full text |Cite
|
Sign up to set email alerts
|

Machine Learning Assisted Clustering of Nanoparticle Structures

Abstract: We propose a scheme for the automatic separation (i.e., clustering) of data sets composed of several nanoparticle (NP) structures by means of Machine Learning techniques. These data sets originate from atomistic simulations, such as global optimizations searches and molecular dynamics simulations, which can produce large outputs that are often difficult to inspect by hand. By combining a description of NPs based on their local atomic environment with unsupervised learning algorithms, such as K-Means and Gaussi… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
2
1

Citation Types

0
9
2

Year Published

2023
2023
2024
2024

Publication Types

Select...
5
1

Relationship

1
5

Authors

Journals

citations
Cited by 13 publications
(11 citation statements)
references
References 45 publications
0
9
2
Order By: Relevance
“…We note that very recently Roncaglia and Ferrando introduced many CNA signatures for Ag, AgCu, Au, and AuPd, which appeared for clusters near the melting point. 86 However, the signatures identified here appear to be different from those in their study. Ag Nanocrystals in Vacuum.…”
Section: Resultscontrasting
confidence: 93%
See 1 more Smart Citation
“…We note that very recently Roncaglia and Ferrando introduced many CNA signatures for Ag, AgCu, Au, and AuPd, which appeared for clusters near the melting point. 86 However, the signatures identified here appear to be different from those in their study. Ag Nanocrystals in Vacuum.…”
Section: Resultscontrasting
confidence: 93%
“…Figure f–i depicts each of these atom types, and their CNA designations are included in Table , where they are shaded in gray. We note that very recently Roncaglia and Ferrando introduced many CNA signatures for Ag, AgCu, Au, and AuPd, which appeared for clusters near the melting point . However, the signatures identified here appear to be different from those in their study.…”
Section: Resultscontrasting
confidence: 64%
“…27 The authors note that more extensive atom classification schemes exist in the literature, such as a recently published 63-class scheme which underpins a machine-learning cluster classification system. 28 However, the simpler method used herein is appropriate for the scope of this work as the classification is used to broadly understand the nature of the different funnels, rather than perform detailed structural analysis.…”
Section: ■ Methodsmentioning
confidence: 99%
“…In this scheme, clusters are classified as icosahedral (ICO), decahedral (DEC), face-centered cubic (FCC), hexagonal close-packed (HCP), or ambiguous (AMB) based on an underlying atom classification method with 12 atom classes . The authors note that more extensive atom classification schemes exist in the literature, such as a recently published 63-class scheme which underpins a machine-learning cluster classification system . However, the simpler method used herein is appropriate for the scope of this work as the classification is used to broadly understand the nature of the different funnels, rather than perform detailed structural analysis.…”
Section: Methodsmentioning
confidence: 99%
“…First, CNA classifications are based on the arrangement of first neighbors around any given atom, and therefore, they do not directly encode information on the overall shape of the nanoparticles. In addition, even though CNA can be used for charting the structural landscape and for unsupervised clustering to obtain very refined groupings of structures (e.g., along the lines developed by Roncaglia and Ferrando 27 ), the resulting chart is nondifferentiable.…”
Section: Introductionmentioning
confidence: 99%