Visualizing Temporal Topic Embeddings with a Compass

Dynamic topic modeling is useful at discovering the development and change in latent topics over time. However, present methodology relies on algorithms that separate document and word representations. This prevents the creation of a meaningful embedding space where changes in word usage and documen...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:IEEE transactions on visualization and computer graphics Jg. 31; H. 1; S. 272 - 282
Hauptverfasser: Palamarchuk, Daniel, Williams, Lemara, Mayer, Brian, Danielson, Thomas, Faust, Rebecca, Deschaine, Larry, North, Chris
Format: Journal Article
Sprache:Englisch
Veröffentlicht: United States IEEE 01.01.2025
Schlagworte:
ISSN:1077-2626, 1941-0506, 1941-0506
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Abstract Dynamic topic modeling is useful at discovering the development and change in latent topics over time. However, present methodology relies on algorithms that separate document and word representations. This prevents the creation of a meaningful embedding space where changes in word usage and documents can be directly analyzed in a temporal context. This paper proposes an expansion of the compass-aligned temporal Word2Vec methodology into dynamic topic modeling. Such a method allows for the direct comparison of word and document embeddings across time in dynamic topics. This enables the creation of visualizations that incorporate temporal word embeddings within the context of documents into topic visualizations. In experiments against the current state-of-the-art, our proposed method demonstrates overall competitive performance in topic relevancy and diversity across temporal datasets of varying size. Simultaneously, it provides insightful visualizations focused on temporal word embeddings while maintaining the insights provided by global topic evolution, advancing our understanding of how topics evolve over time.
AbstractList Dynamic topic modeling is useful at discovering the development and change in latent topics over time. However, present methodology relies on algorithms that separate document and word representations. This prevents the creation of a meaningful embedding space where changes in word usage and documents can be directly analyzed in a temporal context. This paper proposes an expansion of the compass-aligned temporal Word2Vec methodology into dynamic topic modeling. Such a method allows for the direct comparison of word and document embeddings across time in dynamic topics. This enables the creation of visualizations that incorporate temporal word embeddings within the context of documents into topic visualizations. In experiments against the current state-of-the-art, our proposed method demonstrates overall competitive performance in topic relevancy and diversity across temporal datasets of varying size. Simultaneously, it provides insightful visualizations focused on temporal word embeddings while maintaining the insights provided by global topic evolution, advancing our understanding of how topics evolve over time.
Dynamic topic modeling is useful at discovering the development and change in latent topics over time. However, present methodology relies on algorithms that separate document and word representations. This prevents the creation of a meaningful embedding space where changes in word usage and documents can be directly analyzed in a temporal context. This paper proposes an expansion of the compass-aligned temporal Word2Vec methodology into dynamic topic modeling. Such a method allows for the direct comparison of word and document embeddings across time in dynamic topics. This enables the creation of visualizations that incorporate temporal word embeddings within the context of documents into topic visualizations. In experiments against the current state-of-the-art, our proposed method demonstrates overall competitive performance in topic relevancy and diversity across temporal datasets of varying size. Simultaneously, it provides insightful visualizations focused on temporal word embeddings while maintaining the insights provided by global topic evolution, advancing our understanding of how topics evolve over time.Dynamic topic modeling is useful at discovering the development and change in latent topics over time. However, present methodology relies on algorithms that separate document and word representations. This prevents the creation of a meaningful embedding space where changes in word usage and documents can be directly analyzed in a temporal context. This paper proposes an expansion of the compass-aligned temporal Word2Vec methodology into dynamic topic modeling. Such a method allows for the direct comparison of word and document embeddings across time in dynamic topics. This enables the creation of visualizations that incorporate temporal word embeddings within the context of documents into topic visualizations. In experiments against the current state-of-the-art, our proposed method demonstrates overall competitive performance in topic relevancy and diversity across temporal datasets of varying size. Simultaneously, it provides insightful visualizations focused on temporal word embeddings while maintaining the insights provided by global topic evolution, advancing our understanding of how topics evolve over time.
Author Mayer, Brian
North, Chris
Faust, Rebecca
Danielson, Thomas
Deschaine, Larry
Palamarchuk, Daniel
Williams, Lemara
Author_xml – sequence: 1
  givenname: Daniel
  surname: Palamarchuk
  fullname: Palamarchuk, Daniel
  email: d4n1elp@vt.edu
  organization: Virginia Polytechnic Institute and State University, Blacksburg, VA, USA
– sequence: 2
  givenname: Lemara
  surname: Williams
  fullname: Williams, Lemara
  email: lemaraw@vt.edu
  organization: Virginia Polytechnic Institute and State University, Blacksburg, VA, USA
– sequence: 3
  givenname: Brian
  surname: Mayer
  fullname: Mayer, Brian
  email: bmayer@cs.vt.edu
  organization: Virginia Polytechnic Institute and State University, Blacksburg, VA, USA
– sequence: 4
  givenname: Thomas
  surname: Danielson
  fullname: Danielson, Thomas
  email: Thomas.Danielson@srnl.doe.gov
  organization: Savannah River National Laboratory, USA
– sequence: 5
  givenname: Rebecca
  surname: Faust
  fullname: Faust, Rebecca
  email: rfaust1@tulane.edu
  organization: Tulane University, USA
– sequence: 6
  givenname: Larry
  surname: Deschaine
  fullname: Deschaine, Larry
  email: larry.deschaine@srnl.doe.gov
  organization: Savannah River National Laboratory, USA
– sequence: 7
  givenname: Chris
  orcidid: 0000-0002-8786-7103
  surname: North
  fullname: North, Chris
  email: north@cs.vt.edu
  organization: Virginia Polytechnic Institute and State University, Blacksburg, VA, USA
BackLink https://www.ncbi.nlm.nih.gov/pubmed/39255128$$D View this record in MEDLINE/PubMed
BookMark eNpNkE1Lw0AQhhdR7If-AEEkRy-pM_uZHKXUKhS8xF6XzWajK_ky2yD6601pFU8zwzzvwDwzctq0jSPkCmGBCOldtl2uFxQoXzAuJHJ2QqaYcoxBgDwde1AqppLKCZmF8A6AnCfpOZmwlAqBNJkSvvVhMJX_9s1rlLm6a3tTRVnbeRut6twVxbgI0affvUUmWrZ1Z0K4IGelqYK7PNY5eXlYZcvHePO8flreb2LLAHexASvyXHEURkqUNncqEawAWXBmISkpjGMJTFKVMmZKUTgobVqwnJaoOGdzcnu42_Xtx-DCTtc-WFdVpnHtEDRDoImSMmEjenNEh7x2he56X5v-S_9-OgJ4AGzfhtC78g9B0Hubem9T723qo80xc33IeOfcP14q4CjZD6tibfc
CODEN ITVGEA
Cites_doi 10.1145/3178876.3185999
10.1111/j.1467-8659.2012.03108.x
10.1162/tacl_a_00325
10.1609/aaai.v33i01.33016326
10.1177/2053168017712821
10.1109/TVCG.2016.2515592
10.1145/3159652.3159703
10.1109/TVCG.2014.2346433
10.18653/v1/P16-1141
10.18653/v1/P17-2071
10.1109/TVCG.2013.221
10.1145/2089094.2089101
10.1080/00437956.1954.11659520
10.1109/VAST.2016.7883511
10.1007/s13222-013-0134-x
10.18653/v1/D19-1410
10.1145/1143844.1143859
10.1109/TVCG.2013.162
10.1145/1081870.1081895
10.3115/v1/W14-3110
10.1109/tvcg.2011.239
10.1109/2945.981848
10.1109/TVCG.2015.2467618
10.1109/TVCG.2014.2388208
10.1145/2492517.2492639
10.1109/TVCG.2014.2346919
10.1145/2684822.2685324
10.1007/978-3-642-37456-2_14
10.1007/s12650-015-0323-9
ContentType Journal Article
DBID 97E
RIA
RIE
AAYXX
CITATION
NPM
7X8
DOI 10.1109/TVCG.2024.3456143
DatabaseName IEEE All-Society Periodicals Package (ASPP) 2005-present
IEEE All-Society Periodicals Package (ASPP) 1998-Present
IEEE Electronic Library (IEL)
CrossRef
PubMed
MEDLINE - Academic
DatabaseTitle CrossRef
PubMed
MEDLINE - Academic
DatabaseTitleList
MEDLINE - Academic
PubMed
Database_xml – sequence: 1
  dbid: NPM
  name: PubMed
  url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed
  sourceTypes: Index Database
– sequence: 2
  dbid: RIE
  name: IEEE Xplore Digital Library (LUT)
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
– sequence: 3
  dbid: 7X8
  name: MEDLINE - Academic
  url: https://search.proquest.com/medline
  sourceTypes: Aggregation Database
DeliveryMethod fulltext_linktorsrc
Discipline Engineering
EISSN 1941-0506
EndPage 282
ExternalDocumentID 39255128
10_1109_TVCG_2024_3456143
10670416
Genre orig-research
Journal Article
GrantInformation_xml – fundername: National Science Foundation
  grantid: 2127309
  funderid: 10.13039/100000001
GroupedDBID ---
-~X
.DC
0R~
29I
4.4
53G
5GY
5VS
6IK
97E
AAJGR
AARMG
AASAJ
AAWTH
ABAZT
ABQJQ
ABVLG
ACGFO
ACIWK
AENEX
AETIX
AGQYO
AGSQL
AHBIQ
AI.
AIBXA
AKJIK
AKQYR
ALLEH
ALMA_UNASSIGNED_HOLDINGS
ATWAV
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
CS3
DU5
EBS
EJD
F5P
HZ~
H~9
IEDLZ
IFIPE
IFJZH
IPLJI
JAVBF
LAI
M43
O9-
OCL
P2P
PQQKQ
RIA
RIE
RNI
RNS
RZB
TN5
VH1
AAYXX
CITATION
AAYOK
NPM
RIG
7X8
ID FETCH-LOGICAL-c301t-a0c5bb7415a6616cbe7853d06d43c08f20853f03627933af5de0fc9d3b2f17443
IEDL.DBID RIE
ISICitedReferencesCount 0
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001367808800011&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 1077-2626
1941-0506
IngestDate Sun Nov 09 13:18:16 EST 2025
Wed Mar 05 02:44:39 EST 2025
Sat Nov 29 03:31:50 EST 2025
Wed Aug 27 03:03:27 EDT 2025
IsDoiOpenAccess false
IsOpenAccess true
IsPeerReviewed true
IsScholarly true
Issue 1
Language English
License https://ieeexplore.ieee.org/Xplorehelp/downloads/license-information/IEEE.html
https://doi.org/10.15223/policy-029
https://doi.org/10.15223/policy-037
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c301t-a0c5bb7415a6616cbe7853d06d43c08f20853f03627933af5de0fc9d3b2f17443
Notes ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ORCID 0000-0002-8786-7103
OpenAccessLink https://www.osti.gov/servlets/purl/2580486
PMID 39255128
PQID 3102876683
PQPubID 23479
PageCount 11
ParticipantIDs crossref_primary_10_1109_TVCG_2024_3456143
ieee_primary_10670416
proquest_miscellaneous_3102876683
pubmed_primary_39255128
PublicationCentury 2000
PublicationDate 2025-01-01
PublicationDateYYYYMMDD 2025-01-01
PublicationDate_xml – month: 01
  year: 2025
  text: 2025-01-01
  day: 01
PublicationDecade 2020
PublicationPlace United States
PublicationPlace_xml – name: United States
PublicationTitle IEEE transactions on visualization and computer graphics
PublicationTitleAbbrev TVCG
PublicationTitleAlternate IEEE Trans Vis Comput Graph
PublicationYear 2025
Publisher IEEE
Publisher_xml – name: IEEE
References McInnes (ref27) 2020
ref35
ref12
ref34
ref15
ref37
ref14
ref36
ref31
Grootendorst (ref17) 2022
ref30
ref11
ref33
ref10
ref32
ref1
Dieng (ref13) 2019
ref39
ref16
ref38
ref19
Le (ref23) 2014
ref18
Blei (ref6) 2003; 3
Bianchi (ref4) 2020
ref24
ref26
ref25
ref20
ref22
ref21
Mikolov (ref29) 2013
Brown (ref8)
Bouma (ref7) 2009
ref28
ref9
ref3
ref5
Angelov (ref2) 2020
References_xml – ident: ref32
  doi: 10.1145/3178876.3185999
– ident: ref24
  doi: 10.1111/j.1467-8659.2012.03108.x
– year: 2013
  ident: ref29
  publication-title: Efficient estimation of word representations in vector space
– ident: ref14
  doi: 10.1162/tacl_a_00325
– ident: ref12
  doi: 10.1609/aaai.v33i01.33016326
– ident: ref3
  doi: 10.1177/2053168017712821
– ident: ref37
  doi: 10.1109/TVCG.2016.2515592
– ident: ref39
  doi: 10.1145/3159652.3159703
– ident: ref11
  doi: 10.1109/TVCG.2014.2346433
– ident: ref19
  doi: 10.18653/v1/P16-1141
– year: 2020
  ident: ref4
  article-title: Compass-aligned distributional embeddings for studying semantic differences across corpora
  publication-title: arXiv preprint
– ident: ref35
  doi: 10.18653/v1/P17-2071
– ident: ref38
  doi: 10.1109/TVCG.2013.221
– ident: ref25
  doi: 10.1145/2089094.2089101
– ident: ref20
  doi: 10.1080/00437956.1954.11659520
– ident: ref36
  doi: 10.1109/VAST.2016.7883511
– ident: ref18
  doi: 10.1007/s13222-013-0134-x
– ident: ref30
  doi: 10.18653/v1/D19-1410
– volume-title: Top2vec: Distributed representations of topics
  year: 2020
  ident: ref2
– ident: ref5
  doi: 10.1145/1143844.1143859
– year: 2014
  ident: ref23
  publication-title: Distributed representations of sentences and documents
– ident: ref15
  doi: 10.1109/TVCG.2013.162
– ident: ref28
  doi: 10.1145/1081870.1081895
– year: 2019
  ident: ref13
  publication-title: The dynamic embedded topic model
– volume: 3
  start-page: 993
  year: 2003
  ident: ref6
  article-title: Latent dirichlet allocation
  publication-title: J. Mach. Learn. Res.
– volume-title: Proceedings of the 34th International Conference on Neural Information Processing Systems, NIPS’20
  ident: ref8
  article-title: Language models are few-shot learners
– ident: ref33
  doi: 10.3115/v1/W14-3110
– ident: ref10
  doi: 10.1109/tvcg.2011.239
– ident: ref21
  doi: 10.1109/2945.981848
– ident: ref1
  doi: 10.1109/TVCG.2015.2467618
– year: 2009
  ident: ref7
  publication-title: Normalized (pointwise) mutual information in collocation extraction
– year: 2020
  ident: ref27
  publication-title: Umap: Uniform manifold approximation and projection for dimension reduction
– ident: ref16
  doi: 10.1109/TVCG.2014.2388208
– ident: ref26
  doi: 10.1145/2492517.2492639
– ident: ref34
  doi: 10.1109/TVCG.2014.2346919
– ident: ref31
  doi: 10.1145/2684822.2685324
– ident: ref9
  doi: 10.1007/978-3-642-37456-2_14
– ident: ref22
  doi: 10.1007/s12650-015-0323-9
– year: 2022
  ident: ref17
  publication-title: Bertopic: Neural topic modeling with a class-based tf-idf procedure
SSID ssj0014489
Score 2.438974
Snippet Dynamic topic modeling is useful at discovering the development and change in latent topics over time. However, present methodology relies on algorithms that...
SourceID proquest
pubmed
crossref
ieee
SourceType Aggregation Database
Index Database
Publisher
StartPage 272
SubjectTerms Cluster analysis
Clustering algorithms
Compass
Dynamic topic modeling
Electronic mail
High dimensional data
Streams
Training
Vectors
Visualization
Title Visualizing Temporal Topic Embeddings with a Compass
URI https://ieeexplore.ieee.org/document/10670416
https://www.ncbi.nlm.nih.gov/pubmed/39255128
https://www.proquest.com/docview/3102876683
Volume 31
WOSCitedRecordID wos001367808800011&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
journalDatabaseRights – providerCode: PRVIEE
  databaseName: IEEE Xplore Digital Library (LUT)
  customDbUrl:
  eissn: 1941-0506
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0014489
  issn: 1077-2626
  databaseCode: RIE
  dateStart: 19950101
  isFulltext: true
  titleUrlDefault: https://ieeexplore.ieee.org/
  providerName: IEEE
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV09T8MwELWgYoCBzwLlowoSE1JaN3bieERVC1PFEKpuUezYUiVIqqZl4Ndz56RVlw5siZREyfM5987nu0fIs4GRNFYOYCKpwOdgNL7SlvmRlMJkImOhtU5sQkwm8WwmP5pidVcLY4xxm89MDw9dLj8v9RqXyvrY7owCgzgkh0KIulhrmzKAOEPWGwyFHwBNb1KYAyr7yXT4BqFgwHsM-QJH8RzgBUAWUIR9xx85gZX9XNP5nPHZP9_2nJw25NJ7ra3hghyY4pKc7LQcvCJ8Oq-wjvIXzryk7kv15SXlYq690bcyuctFebg862We-1tUVZt8jkfJ8N1vlBN8DRN25WdUh0ohWcjA_0ZaGQFuOadRzpmmsUVhTmbRecH0ZJkNc0OtljlTgYUQhbNr0irKwtwST4uQGxvCh2jw5iKXVCplhaWDnGcRVx3yssEvXdQNMlIXWFCZIu4p4p42uHdIG3HaubCGqEOeNpCnYN6Ys8gKU66rlCEBElEUw7039Vhs794M4d2ep96T4wDVet2CyQNprZZr80iO9M9qXi27YEOzuOts6A-MR77M
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV09T8MwED1BQQIGPguUzyAxIaV1YyeuR4SAIqBiCIgtih1bqgQtIi0Dv547J0VdGNgSKYmS53Punc93D-Dc4khap7o4kXQUCjSaUBvHw0QpaXOZ89g5LzYhB4Pe66t6qovVfS2MtdZvPrNtOvS5_GJsprRU1qF2ZwwZxCIsxUJE3apc6zdpgJGGqrYYyjBCol4nMbtMddKXq1sMBiPR5sQYBMnnIDNAukAy7HMeyUus_M02vde52fjn-27Cek0vg8vKHrZgwY62YW2u6eAOiJdhSZWU33gWpFVnqrcgHX8MTXD9rm3hs1EBLdAGeeD_F2XZhOeb6_SqH9baCaHBKTsJc2ZirYku5OiBE6OtRMdcsKQQ3LCeI2lO7sh94QTluYsLy5xRBdeRwyBF8F1ojMYjuw-BkbGwLsYPMejPZaGY0tpJx7qFyBOhW3Axwy_7qFpkZD60YCoj3DPCPatxb0GTcJq7sIKoBWczyDM0cMpa5CM7npYZJwokk6SH9-5VY_F792wID_546ims9NPHh-zhbnB_CKsRaff65ZMjaEw-p_YYls3XZFh-nnhL-gEpjMEr
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Visualizing+Temporal+Topic+Embeddings+with+a+Compass&rft.jtitle=IEEE+transactions+on+visualization+and+computer+graphics&rft.au=Palamarchuk%2C+Daniel&rft.au=Williams%2C+Lemara&rft.au=Mayer%2C+Brian&rft.au=Danielson%2C+Thomas&rft.date=2025-01-01&rft.pub=IEEE&rft.issn=1077-2626&rft.volume=31&rft.issue=1&rft.spage=272&rft.epage=282&rft_id=info:doi/10.1109%2FTVCG.2024.3456143&rft_id=info%3Apmid%2F39255128&rft.externalDocID=10670416
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1077-2626&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1077-2626&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1077-2626&client=summon