Evaluation of the proposed algorithm across varying agent counts (1,000, 3,000, and 5,000), using four metrics: Adjusted Rand Index (ARI), Normalized Mutual Information (NMI), Silhouette Score, and Davies-Bouldin Index.

<p>Each metric is plotted against the number of agents. The proposed method demonstrates consistently superior performance compared to the baseline distributed approach. A slight degradation in clustering quality at 5,000 agents indicates potential scalability limitations, which may be address...

Full description

Saved in:
Bibliographic Details
Main Author: Victoria Erofeeva (21807723) (author)
Other Authors: Oleg Granichin (21807726) (author), Vikentii Pankov (21807729) (author), Zeev Volkovich (9611574) (author)
Published: 2025
Subjects:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1852018114463531008
author Victoria Erofeeva (21807723)
author2 Oleg Granichin (21807726)
Vikentii Pankov (21807729)
Zeev Volkovich (9611574)
author2_role author
author
author
author_facet Victoria Erofeeva (21807723)
Oleg Granichin (21807726)
Vikentii Pankov (21807729)
Zeev Volkovich (9611574)
author_role author
dc.creator.none.fl_str_mv Victoria Erofeeva (21807723)
Oleg Granichin (21807726)
Vikentii Pankov (21807729)
Zeev Volkovich (9611574)
dc.date.none.fl_str_mv 2025-07-29T17:34:21Z
dc.identifier.none.fl_str_mv 10.1371/journal.pone.0327396.g003
dc.relation.none.fl_str_mv https://figshare.com/articles/figure/Evaluation_of_the_proposed_algorithm_across_varying_agent_counts_1_000_3_000_and_5_000_using_four_metrics_Adjusted_Rand_Index_ARI_Normalized_Mutual_Information_NMI_Silhouette_Score_and_Davies-Bouldin_Index_/29667957
dc.rights.none.fl_str_mv CC BY 4.0
info:eu-repo/semantics/openAccess
dc.subject.none.fl_str_mv Cell Biology
Infectious Diseases
Biological Sciences not elsewhere classified
Mathematical Sciences not elsewhere classified
Information Systems not elsewhere classified
world datasets show
unlike traditional methods
trained neural network
dynamic network changes
system &# 8217
minimal communication overhead
efficient decentralized clustering
system adapts
clustering structure
clustering accuracy
xlink ">
static topologies
paper presents
iot ).
immediate results
highly suitable
generate compact
fly processing
dynamical multi
distributed environments
distributed aggregation
dimensionality reduction
data must
consistent summaries
consensus protocol
centralized computation
dc.title.none.fl_str_mv Evaluation of the proposed algorithm across varying agent counts (1,000, 3,000, and 5,000), using four metrics: Adjusted Rand Index (ARI), Normalized Mutual Information (NMI), Silhouette Score, and Davies-Bouldin Index.
dc.type.none.fl_str_mv Image
Figure
info:eu-repo/semantics/publishedVersion
image
description <p>Each metric is plotted against the number of agents. The proposed method demonstrates consistently superior performance compared to the baseline distributed approach. A slight degradation in clustering quality at 5,000 agents indicates potential scalability limitations, which may be addressed through architectural modifications such as increased network depth or additional internal channels.</p>
eu_rights_str_mv openAccess
id Manara_787056969ab09fa026adfac97684ba4e
identifier_str_mv 10.1371/journal.pone.0327396.g003
network_acronym_str Manara
network_name_str ManaraRepo
oai_identifier_str oai:figshare.com:article/29667957
publishDate 2025
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
rights_invalid_str_mv CC BY 4.0
spelling Evaluation of the proposed algorithm across varying agent counts (1,000, 3,000, and 5,000), using four metrics: Adjusted Rand Index (ARI), Normalized Mutual Information (NMI), Silhouette Score, and Davies-Bouldin Index.Victoria Erofeeva (21807723)Oleg Granichin (21807726)Vikentii Pankov (21807729)Zeev Volkovich (9611574)Cell BiologyInfectious DiseasesBiological Sciences not elsewhere classifiedMathematical Sciences not elsewhere classifiedInformation Systems not elsewhere classifiedworld datasets showunlike traditional methodstrained neural networkdynamic network changessystem &# 8217minimal communication overheadefficient decentralized clusteringsystem adaptsclustering structureclustering accuracyxlink ">static topologiespaper presentsiot ).immediate resultshighly suitablegenerate compactfly processingdynamical multidistributed environmentsdistributed aggregationdimensionality reductiondata mustconsistent summariesconsensus protocolcentralized computation<p>Each metric is plotted against the number of agents. The proposed method demonstrates consistently superior performance compared to the baseline distributed approach. A slight degradation in clustering quality at 5,000 agents indicates potential scalability limitations, which may be addressed through architectural modifications such as increased network depth or additional internal channels.</p>2025-07-29T17:34:21ZImageFigureinfo:eu-repo/semantics/publishedVersionimage10.1371/journal.pone.0327396.g003https://figshare.com/articles/figure/Evaluation_of_the_proposed_algorithm_across_varying_agent_counts_1_000_3_000_and_5_000_using_four_metrics_Adjusted_Rand_Index_ARI_Normalized_Mutual_Information_NMI_Silhouette_Score_and_Davies-Bouldin_Index_/29667957CC BY 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/296679572025-07-29T17:34:21Z
spellingShingle Evaluation of the proposed algorithm across varying agent counts (1,000, 3,000, and 5,000), using four metrics: Adjusted Rand Index (ARI), Normalized Mutual Information (NMI), Silhouette Score, and Davies-Bouldin Index.
Victoria Erofeeva (21807723)
Cell Biology
Infectious Diseases
Biological Sciences not elsewhere classified
Mathematical Sciences not elsewhere classified
Information Systems not elsewhere classified
world datasets show
unlike traditional methods
trained neural network
dynamic network changes
system &# 8217
minimal communication overhead
efficient decentralized clustering
system adapts
clustering structure
clustering accuracy
xlink ">
static topologies
paper presents
iot ).
immediate results
highly suitable
generate compact
fly processing
dynamical multi
distributed environments
distributed aggregation
dimensionality reduction
data must
consistent summaries
consensus protocol
centralized computation
status_str publishedVersion
title Evaluation of the proposed algorithm across varying agent counts (1,000, 3,000, and 5,000), using four metrics: Adjusted Rand Index (ARI), Normalized Mutual Information (NMI), Silhouette Score, and Davies-Bouldin Index.
title_full Evaluation of the proposed algorithm across varying agent counts (1,000, 3,000, and 5,000), using four metrics: Adjusted Rand Index (ARI), Normalized Mutual Information (NMI), Silhouette Score, and Davies-Bouldin Index.
title_fullStr Evaluation of the proposed algorithm across varying agent counts (1,000, 3,000, and 5,000), using four metrics: Adjusted Rand Index (ARI), Normalized Mutual Information (NMI), Silhouette Score, and Davies-Bouldin Index.
title_full_unstemmed Evaluation of the proposed algorithm across varying agent counts (1,000, 3,000, and 5,000), using four metrics: Adjusted Rand Index (ARI), Normalized Mutual Information (NMI), Silhouette Score, and Davies-Bouldin Index.
title_short Evaluation of the proposed algorithm across varying agent counts (1,000, 3,000, and 5,000), using four metrics: Adjusted Rand Index (ARI), Normalized Mutual Information (NMI), Silhouette Score, and Davies-Bouldin Index.
title_sort Evaluation of the proposed algorithm across varying agent counts (1,000, 3,000, and 5,000), using four metrics: Adjusted Rand Index (ARI), Normalized Mutual Information (NMI), Silhouette Score, and Davies-Bouldin Index.
topic Cell Biology
Infectious Diseases
Biological Sciences not elsewhere classified
Mathematical Sciences not elsewhere classified
Information Systems not elsewhere classified
world datasets show
unlike traditional methods
trained neural network
dynamic network changes
system &# 8217
minimal communication overhead
efficient decentralized clustering
system adapts
clustering structure
clustering accuracy
xlink ">
static topologies
paper presents
iot ).
immediate results
highly suitable
generate compact
fly processing
dynamical multi
distributed environments
distributed aggregation
dimensionality reduction
data must
consistent summaries
consensus protocol
centralized computation