Investigating antiquities trafficking with generative pre-trained transformer (GPT)-3 enabled knowledge graphs: A case study

Background: There is a wide variety of potential sources from which insight into the antiquities trade could be culled, from newspaper articles to auction catalogues, to court dockets, to personal archives, if it could all be systematically examined. We explore the use of a large language model, GPT...

Full description

Saved in:
Bibliographic Details
Published in:Open research Europe Vol. 3; p. 100
Main Authors: Graham, Shawn, Yates, Donna, El-Roby, Ahmed
Format: Journal Article
Language:English
Published: London, UK F1000 Research Limited 2023
F1000 Research Ltd
Subjects:
ISSN:2732-5121, 2732-5121
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Abstract Background: There is a wide variety of potential sources from which insight into the antiquities trade could be culled, from newspaper articles to auction catalogues, to court dockets, to personal archives, if it could all be systematically examined. We explore the use of a large language model, GPT-3, to semi-automate the creation of a knowledge graph of a body of scholarship concerning the antiquities trade. Methods: We give GPT-3 a prompt guiding it to identify knowledge statements around the trade. Given GPT-3’s understanding of the statistical properties of language, our prompt teaches GPT-3 to append text to each article we feed it where the appended text summarizes the knowledge in the article. The summary is in the form of a list of subject, predicate, and object relationships, representing a knowledge graph. Previously we created such lists by manually annotating the source articles. We compare the result of this automatic process with a knowledge graph created from the same sources via hand. When such knowledge graphs are projected into a multi-dimensional embedding model using a neural network (via the Ampligraph open-source Python library), the relative positioning of entities implies the probability of a connection; the direction of the positioning implies the kind of connection. Thus, we can interrogate the embedding model to discover new probable relationships. The results can generate new insight about the antiquity trade, suggesting possible avenues of research. Results: We find that our semi-automatic approach to generating the knowledge graph in the first place produces comparable results to our hand-made version, but at an enormous savings of time and a possible expansion of the amount of materials we can consider. Conclusions: These results have implications for working with other kinds of archaeological knowledge in grey literature, reports, articles, and other venues via computational means.
AbstractList Background: There is a wide variety of potential sources from which insight into the antiquities trade could be culled, from newspaper articles to auction catalogues, to court dockets, to personal archives, if it could all be systematically examined. We explore the use of a large language model, GPT-3, to semi-automate the creation of a knowledge graph of a body of scholarship concerning the antiquities trade. Methods: We give GPT-3 a prompt guiding it to identify knowledge statements around the trade. Given GPT-3's understanding of the statistical properties of language, our prompt teaches GPT-3 to append text to each article we feed it where the appended text summarizes the knowledge in the article. The summary is in the form of a list of subject, predicate, and object relationships, representing a knowledge graph. Previously we created such lists by manually annotating the source articles. We compare the result of this automatic process with a knowledge graph created from the same sources via hand. When such knowledge graphs are projected into a multi-dimensional embedding model using a neural network (via the Ampligraph open-source Python library), the relative positioning of entities implies the probability of a connection; the direction of the positioning implies the kind of connection. Thus, we can interrogate the embedding model to discover new probable relationships. The results can generate new insight about the antiquity trade, suggesting possible avenues of research. Results: We find that our semi-automatic approach to generating the knowledge graph in the first place produces comparable results to our hand-made version, but at an enormous savings of time and a possible expansion of the amount of materials we can consider. Conclusions: These results have implications for working with other kinds of archaeological knowledge in grey literature, reports, articles, and other venues via computational means.Background: There is a wide variety of potential sources from which insight into the antiquities trade could be culled, from newspaper articles to auction catalogues, to court dockets, to personal archives, if it could all be systematically examined. We explore the use of a large language model, GPT-3, to semi-automate the creation of a knowledge graph of a body of scholarship concerning the antiquities trade. Methods: We give GPT-3 a prompt guiding it to identify knowledge statements around the trade. Given GPT-3's understanding of the statistical properties of language, our prompt teaches GPT-3 to append text to each article we feed it where the appended text summarizes the knowledge in the article. The summary is in the form of a list of subject, predicate, and object relationships, representing a knowledge graph. Previously we created such lists by manually annotating the source articles. We compare the result of this automatic process with a knowledge graph created from the same sources via hand. When such knowledge graphs are projected into a multi-dimensional embedding model using a neural network (via the Ampligraph open-source Python library), the relative positioning of entities implies the probability of a connection; the direction of the positioning implies the kind of connection. Thus, we can interrogate the embedding model to discover new probable relationships. The results can generate new insight about the antiquity trade, suggesting possible avenues of research. Results: We find that our semi-automatic approach to generating the knowledge graph in the first place produces comparable results to our hand-made version, but at an enormous savings of time and a possible expansion of the amount of materials we can consider. Conclusions: These results have implications for working with other kinds of archaeological knowledge in grey literature, reports, articles, and other venues via computational means.
Background: There is a wide variety of potential sources from which insight into the antiquities trade could be culled, from newspaper articles to auction catalogues, to court dockets, to personal archives, if it could all be systematically examined. We explore the use of a large language model, GPT-3, to semi-automate the creation of a knowledge graph of a body of scholarship concerning the antiquities trade. Methods: We give GPT-3 a prompt guiding it to identify knowledge statements around the trade. Given GPT-3’s understanding of the statistical properties of language, our prompt teaches GPT-3 to append text to each article we feed it where the appended text summarizes the knowledge in the article. The summary is in the form of a list of subject, predicate, and object relationships, representing a knowledge graph. Previously we created such lists by manually annotating the source articles. We compare the result of this automatic process with a knowledge graph created from the same sources via hand. When such knowledge graphs are projected into a multi-dimensional embedding model using a neural network (via the Ampligraph open-source Python library), the relative positioning of entities implies the probability of a connection; the direction of the positioning implies the kind of connection. Thus, we can interrogate the embedding model to discover new probable relationships. The results can generate new insight about the antiquity trade, suggesting possible avenues of research. Results: We find that our semi-automatic approach to generating the knowledge graph in the first place produces comparable results to our hand-made version, but at an enormous savings of time and a possible expansion of the amount of materials we can consider. Conclusions: These results have implications for working with other kinds of archaeological knowledge in grey literature, reports, articles, and other venues via computational means.
Background: There is a wide variety of potential sources from which insight into the antiquities trade could be culled, from newspaper articles to auction catalogues, to court dockets, to personal archives, if it could all be systematically examined. We explore the use of a large language model, GPT-3, to semi-automate the creation of a knowledge graph of a body of scholarship concerning the antiquities trade. Methods: We give GPT-3 a prompt guiding it to identify knowledge statements around the trade. Given GPT-3’s understanding of the statistical properties of language, our prompt teaches GPT-3 to append text to each article we feed it where the appended text summarizes the knowledge in the article. The summary is in the form of a list of subject, predicate, and object relationships, representing a knowledge graph. Previously we created such lists by manually annotating the source articles. We compare the result of this automatic process with a knowledge graph created from the same sources via hand. When such knowledge graphs are projected into a multi-dimensional embedding model using a neural network (via the Ampligraph open-source Python library), the relative positioning of entities implies the probability of a connection; the direction of the positioning implies the kind of connection. Thus, we can interrogate the embedding model to discover new probable relationships. The results can generate new insight about the antiquity trade, suggesting possible avenues of research. Results: We find that our semi-automatic approach to generating the knowledge graph in the first place produces comparable results to our hand-made version, but at an enormous savings of time and a possible expansion of the amount of materials we can consider. Conclusions: These results have implications for working with other kinds of archaeological knowledge in grey literature, reports, articles, and other venues via computational means.
Background: There is a wide variety of potential sources from which insight into the antiquities trade could be culled, from newspaper articles to auction catalogues, to court dockets, to personal archives, if it could all be systematically examined. We explore the use of a large language model, GPT-3, to semi-automate the creation of a knowledge graph of a body of scholarship concerning the antiquities trade. Methods: We give GPT-3 a prompt guiding it to identify knowledge statements around the trade. Given GPT-3’s understanding of the statistical properties of language, our prompt teaches GPT-3 to append text to each article we feed it where the appended text summarizes the knowledge in the article. The summary is in the form of a list of subject, predicate, and object relationships, representing a knowledge graph. Previously we created such lists by manually annotating the source articles. We compare the result of this automatic process with a knowledge graph created from the same sources via hand. When such knowledge graphs are projected into a multi-dimensional embedding model using a neural network (via the Ampligraph open-source Python library), the relative positioning of entities implies the probability of a connection; the direction of the positioning implies the kind of connection. Thus, we can interrogate the embedding model to discover new probable relationships. The results can generate new insight about the antiquity trade, suggesting possible avenues of research. Results: We find that our semi-automatic approach to generating the knowledge graph in the first place produces comparable results to our hand-made version, but at an enormous savings of time and a possible expansion of the amount of materials we can consider. Conclusions: These results have implications for working with other kinds of archaeological knowledge in grey literature, reports, articles, and other venues via computational means.
Author Yates, Donna
Graham, Shawn
El-Roby, Ahmed
Author_xml – sequence: 1
  givenname: Shawn
  surname: Graham
  fullname: Graham, Shawn
– sequence: 2
  givenname: Donna
  orcidid: 0000-0002-9936-6461
  surname: Yates
  fullname: Yates, Donna
– sequence: 3
  givenname: Ahmed
  surname: El-Roby
  fullname: El-Roby, Ahmed
BookMark eNpVkU9r3DAQxUVJoWmar1B0TA_e6p9luZcSQpsuBNpDehayPPIq8UqOZG8I9MNXuxtK9zTDvMdvmHnv0VmIARD6SMmKMqnU5zhBSJBhSaVbUUkIX9E36Jw1nFU1ZfTsv_4dusz5gRDCasolbc_Rn3XYQZ79YGYfBmzC7J8WP3vIeE7GOW8f9_NnP2_wAAFS8e0ATwmqovsA_d4XsotpCwlf3f66_1RxDMF0Y9EeQ3wudQA8JDNt8hd8ja3JgPO89C8f0FtnxgyXr_UC_f7-7f7mR3X383Z9c31XWc4IrQCUcrZhRrSyBV4zW5OOMicaxazreQeNYL2Uoq876lrHhKydU0xQyxvoLL9A6yO3j-ZBT8lvTXrR0Xh9GMQ0aJNmb0fQ1vSMWWFaLp0ASZXowSkB0DVKde2e9fXImpZuC72FUO4fT6CnSvAbPcSdpkSIWhFRCFevhBSflvJ8vfXZwjiaAHHJmqlata2gqi1WebTaFHNO4P7toUQf8tcn-etD_pryv1J8rYw
Cites_doi 10.37911/9781947864382
10.5281/zenodo.6567481
10.1179/0093469013Z.00000000053
10.1086/691725
10.1145/3442188.3445922
10.1017/aap.2023.1
10.1017/S094073911900016X
10.1515/opar-2015-0004
10.48550/arXiv.1810.04805
10.17863/CAM.15975
10.1057/978-1-137-54405-6_10
10.7183/0002-7316.79.1.5
10.48550/arXiv.2211.15661
10.5744/florida/9780813029726.003.0010
10.48550/arXiv.2106.07742
10.48550/arXiv.2302.07736
10.48550/arXiv.2212.03551
10.2307/507226
10.48550/arXiv.2210.13382
10.1007/978-3-030-84856-9_2
10.48550/arXiv.2303.15056
10.4324/9781351026826-8
10.1007/s10611-011-9321-6
ContentType Journal Article
Copyright Copyright: © 2023 Graham S et al.
Copyright: © 2023 Graham S et al. 2023
Copyright_xml – notice: Copyright: © 2023 Graham S et al.
– notice: Copyright: © 2023 Graham S et al. 2023
DBID AAYXX
CITATION
7X8
5PM
DOA
DOI 10.12688/openreseurope.16003.1
DatabaseName CrossRef
MEDLINE - Academic
PubMed Central (Full Participant titles)
DOAJ Directory of Open Access Journals
DatabaseTitle CrossRef
MEDLINE - Academic
DatabaseTitleList MEDLINE - Academic


CrossRef
Database_xml – sequence: 1
  dbid: DOA
  name: DOAJ Directory of Open Access Journals
  url: https://www.doaj.org/
  sourceTypes: Open Website
– sequence: 2
  dbid: 7X8
  name: MEDLINE - Academic
  url: https://search.proquest.com/medline
  sourceTypes: Aggregation Database
DeliveryMethod fulltext_linktorsrc
EISSN 2732-5121
ExternalDocumentID oai_doaj_org_article_cad22c4a936f4e6184def84eeb788b9c
PMC10445804
10_12688_openreseurope_16003_1
GrantInformation_xml – fundername: Horizon 2020 Framework Programme
  grantid: 804851
– fundername: Social Sciences and Humanities Research Council of Canada
GroupedDBID AAFWJ
AAYXX
AFPKN
ALMA_UNASSIGNED_HOLDINGS
CITATION
GROUPED_DOAJ
M~E
OK1
PGMZT
RPM
7X8
5PM
ID FETCH-LOGICAL-c3201-ee88fc72a4969e352c50b12f4782cfd3be742d664d5b1f9f2465ff8241c37ebc3
IEDL.DBID DOA
ISSN 2732-5121
IngestDate Fri Oct 03 12:50:55 EDT 2025
Tue Sep 30 17:13:05 EDT 2025
Fri Jul 11 07:20:44 EDT 2025
Sat Nov 29 05:59:19 EST 2025
IsDoiOpenAccess true
IsOpenAccess true
IsPeerReviewed true
IsScholarly true
Language English
License This is an open access article distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c3201-ee88fc72a4969e352c50b12f4782cfd3be742d664d5b1f9f2465ff8241c37ebc3
Notes ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
No competing interests were disclosed.
ORCID 0000-0002-9936-6461
OpenAccessLink https://doaj.org/article/cad22c4a936f4e6184def84eeb788b9c
PQID 2858994189
PQPubID 23479
ParticipantIDs doaj_primary_oai_doaj_org_article_cad22c4a936f4e6184def84eeb788b9c
pubmedcentral_primary_oai_pubmedcentral_nih_gov_10445804
proquest_miscellaneous_2858994189
crossref_primary_10_12688_openreseurope_16003_1
PublicationCentury 2000
PublicationDate 2023-00-00
PublicationDateYYYYMMDD 2023-01-01
PublicationDate_xml – year: 2023
  text: 2023-00-00
PublicationDecade 2020
PublicationPlace London, UK
PublicationPlace_xml – name: London, UK
PublicationTitle Open research Europe
PublicationYear 2023
Publisher F1000 Research Limited
F1000 Research Ltd
Publisher_xml – name: F1000 Research Limited
– name: F1000 Research Ltd
References M Bogdanos (ref-4) 2021
E Gilgan (ref-14) 2001
K Kintigh (ref-24) 2014; 79
C Chippendale (ref-7) 2000; 104
A Radford (ref-36) 2018
G Jayaraman (ref-21) 2015
D Yates (ref-45) 2006
T Mashberg (ref-29) 2015
N Oosterman (ref-31) 2018
S Paredes Maury (ref-35) 1999
F Huang (ref-19) 2023
(ref-10) 2019
N Brodie (ref-6) 2019; 26
T Davis (ref-8) 2011; 56
J Felch (ref-12) 2011
M Kersel (ref-23) 2006b
M Kersel (ref-22) 2006a
L Weng (ref-44) 2023
(ref-33) 2018
A Sabar (ref-39) 2020
(ref-34) 2020
E Bender (ref-3) 2021
N Oosterman (ref-32) 2019
E Akyürek (ref-1) 2022
P Krishnankutty (ref-26) 2022
M Levine (ref-28) 2013; 38
S Graham (ref-17) 2023
S Beltrametti (ref-2) 2016; 59
S Gopinathan (ref-15) 2021
A Brandsen (ref-5) 2021
J Devlin (ref-9) 2019
K Li (ref-20) 2022
M Shanahan (ref-40) 2022
S Huang (ref-18) 2022
B Muller (ref-30)
M Fabiani (ref-11) 2021
T Kuzman (ref-27)
P Watson (ref-42) 2007
M Woolf (ref-43) 2023
F Gilardi (ref-13) 2023
K Kintigh (ref-25) 2015; 1
S Graham (ref-16) 2023; 11
I Romanowska (ref-38) 2021
T Underwood (ref-41) 2022
D Richman (ref-37) 2022
References_xml – year: 2001
  ident: ref-14
  article-title: Looting and the Market for Maya Objects: A Belizean Perspective.
– year: 2021
  ident: ref-38
  article-title: Agent-based Modeling for Archaeology: Simulating the Complexity of Societies.
  doi: 10.37911/9781947864382
– year: 1999
  ident: ref-35
  article-title: Surviving in the Rainforest: The Realities of Looting in the Rural Villages of El Petén.
– year: 2022
  ident: ref-41
  article-title: Mapping the Latent Spaces of Culture.
  publication-title: Startwords.
  doi: 10.5281/zenodo.6567481
– year: 2007
  ident: ref-42
  article-title: The Medici Conspiracy: The Illicit Journey of Looted Antiquities From Italy's Tomb Raiders to the Words Greatest Museums.
– volume: 38
  start-page: 264-76
  year: 2013
  ident: ref-28
  article-title: Museum Salvage: A Case Study of Mesoamerican Artifacts in Museum Collections and on the Antiquities Market.
  publication-title: J Field Archaeol.
  doi: 10.1179/0093469013Z.00000000053
– volume: 59
  start-page: 913-44
  year: 2016
  ident: ref-2
  article-title: Market Responses to Court Rulings: Evidence from Antiquities Auctions.
  publication-title: J Law Econ.
  doi: 10.1086/691725
– start-page: 610-23
  year: 2021
  ident: ref-3
  article-title: On the Dangers of Stochastic Parrots: Can Language Models Be Too Big?
  doi: 10.1145/3442188.3445922
– volume: 11
  start-page: 126-138
  year: 2023
  ident: ref-16
  article-title: Relationship Prediction in a Knolwedge Graph Embedding Model of the Illicit Antiquities Trade.
  publication-title: Advances in Archaeological Practice.
  doi: 10.1017/aap.2023.1
– volume: 26
  start-page: 265-283
  year: 2019
  ident: ref-6
  article-title: Through a Glass, Darkly: Long-Term Antiquities Auction Data in Context.
  publication-title: Int J Cult Prop.
  doi: 10.1017/S094073911900016X
– year: 2023
  ident: ref-43
  article-title: ChatGPT's API is So Good and Cheap, It Makes Most Text Generating AI Obsolete
– year: 2018
  ident: ref-33
  article-title: Language Models Are Few-Shot Learners
– ident: ref-27
  article-title: ChatGPT: Beginning of an End of Manual Linguistic Data Annotation? Use Case of Automatic Genre Identification.
– year: 2020
  ident: ref-34
  article-title: Improving Language Understanding with Unsupervised Learning
– year: 2021
  ident: ref-15
  article-title: Meet the Amateur Art Sleuths Fighting To Bring Back India’s Looted Cultural Heritage.
  publication-title: Vice.
– volume: 1
  start-page: 96-101
  year: 2015
  ident: ref-25
  article-title: Extracting Information from Archaeological Texts.
  publication-title: Open Archaeology.
  doi: 10.1515/opar-2015-0004
– year: 2019
  ident: ref-9
  article-title: BERT: Pre-Training of Deep Bidirectional Transformers for Language Understanding.
  doi: 10.48550/arXiv.1810.04805
– year: 2006b
  ident: ref-23
  article-title: License to Sell: The Legal Trade of Antiquities in Israel.
  doi: 10.17863/CAM.15975
– ident: ref-30
  article-title: BERT 101 - State Of The Art NLP Model Explained.
– year: 2023
  ident: ref-44
  article-title: Prompt Engineering
– year: 2011
  ident: ref-12
  article-title: Chasing Aphrodite: The Hunt for Looted Antiquities at the World’s Richest Museum.
– year: 2006
  ident: ref-45
  article-title: South America on the Block: The Changing Face of Pre-Columbian Antiquities Auctions in Response to International Law.
– year: 2022
  ident: ref-18
  article-title: Relationship Extraction with GPT-3.
  publication-title: Geek Culture.
– start-page: 213-35
  year: 2019
  ident: ref-32
  article-title: Regional Overviews of the Policing of Art Crime in the European Union.
  publication-title: The Palgrave Handbook on Art Crime.
  doi: 10.1057/978-1-137-54405-6_10
– year: 2020
  ident: ref-39
  article-title: Veritas: A Harvard Professor, a Con Man, and the Gospel of Jesuss Wife.
– volume: 79
  start-page: 5-24
  year: 2014
  ident: ref-24
  article-title: Grand Challenges for Archaeology.
  publication-title: Am Antiq.
  doi: 10.7183/0002-7316.79.1.5
– year: 2022
  ident: ref-1
  article-title: What Learning Algorithm Is In-Context Learning? Investigations with Linear Models.
  doi: 10.48550/arXiv.2211.15661
– start-page: 188-205
  year: 2006a
  ident: ref-22
  article-title: From the Ground to the Buyer: A Market Analysis of the Trade in Illegal Antiquities.
  publication-title: Archaeology, Cultural Heritage, and the Antiquities Trade.
  doi: 10.5744/florida/9780813029726.003.0010
– year: 2022
  ident: ref-37
  article-title: How to create GPT-3 apps in Google Sheets – Free Tutorial.
  publication-title: Richman SEO Training.
– year: 2021
  ident: ref-5
  article-title: Can BERT Dig It? -- Named Entity Recognition for Information Retrieval in the Archaeology Domain.
  doi: 10.48550/arXiv.2106.07742
– year: 2023
  ident: ref-19
  article-title: Is ChatGPT better than Human Annotators? Potential and Limitations of ChatGPT in Explaining Implicit Hate Speech.
  doi: 10.48550/arXiv.2302.07736
– year: 2023
  ident: ref-17
  article-title: XLabCU/gpt3-relationship-extraction-to-kg: first (0.0.1).
  publication-title: Zenodo
– year: 2021
  ident: ref-4
  article-title: In the Matter of A Grand Jury Investigation into a Private New York Antiquities Collector.
– year: 2022
  ident: ref-40
  article-title: Talking About Large Language Models.
  publication-title: arXiv.
  doi: 10.48550/arXiv.2212.03551
– year: 2019
  ident: ref-10
  article-title: Felony Arrest Warrant, People of New York v. Sanjeeve Asokan et Al.
– year: 2015
  ident: ref-21
  article-title: Intrepid Bloggers Try to Retrieve Stolen Sacred Art from around the World.
  publication-title: India Today.
– year: 2022
  ident: ref-26
  article-title: How Stolen Art Detectives India Pride Project Tracked down 265 Artefacts.
  publication-title: The Print.
– volume: 104
  start-page: 463-511
  year: 2000
  ident: ref-7
  article-title: Material Consequences of Contemporary Classical Collecting.
  publication-title: Am J Archaeol.
  doi: 10.2307/507226
– year: 2022
  ident: ref-20
  article-title: Emergent World Representations: Exploring a Sequence Model Trained on a Synthetic Task.
  doi: 10.48550/arXiv.2210.13382
– start-page: 11-28
  year: 2021
  ident: ref-11
  article-title: Transiting Through the Antiquities Market.
  doi: 10.1007/978-3-030-84856-9_2
– year: 2023
  ident: ref-13
  article-title: ChatGPT Outperforms Crowd-Workers for Text-Annotation Tasks.
  doi: 10.48550/arXiv.2303.15056
– year: 2015
  ident: ref-29
  article-title: Antiquities Dealer Leonardo Patterson Faces New Criminal Charges.
  publication-title: The New York Times.
– year: 2018
  ident: ref-31
  article-title: From Canvas to Ashes: Understanding the Implications of the Westfries Museum and Kunsthal Thefts for the Dutch Art World.
  publication-title: Transnational Crime.
  doi: 10.4324/9781351026826-8
– year: 2018
  ident: ref-36
  article-title: Improving language understanding with unsupervised learning.
– volume: 56
  start-page: 155-74
  year: 2011
  ident: ref-8
  article-title: Supply and Demand: Exposing the Illicit Trade in Cambodian Antiquities through a Study of Sotheby’s Auction House.
  publication-title: Crime Law Soc Change.
  doi: 10.1007/s10611-011-9321-6
SSID ssj0002513619
Score 2.2473717
Snippet Background: There is a wide variety of potential sources from which insight into the antiquities trade could be culled, from newspaper articles to auction...
Background: There is a wide variety of potential sources from which insight into the antiquities trade could be culled, from newspaper articles to auction...
SourceID doaj
pubmedcentral
proquest
crossref
SourceType Open Website
Open Access Repository
Aggregation Database
Index Database
StartPage 100
SubjectTerms antiquities trade
eng
gpt3
illicit antiquities
knowledge graph
knowledge graph embedding model
Large language models
Title Investigating antiquities trafficking with generative pre-trained transformer (GPT)-3 enabled knowledge graphs: A case study
URI https://www.proquest.com/docview/2858994189
https://pubmed.ncbi.nlm.nih.gov/PMC10445804
https://doaj.org/article/cad22c4a936f4e6184def84eeb788b9c
Volume 3
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
journalDatabaseRights – providerCode: PRVAON
  databaseName: DOAJ Directory of Open Access Journals
  customDbUrl:
  eissn: 2732-5121
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0002513619
  issn: 2732-5121
  databaseCode: DOA
  dateStart: 20210101
  isFulltext: true
  titleUrlDefault: https://www.doaj.org/
  providerName: Directory of Open Access Journals
– providerCode: PRVHPJ
  databaseName: ROAD: Directory of Open Access Scholarly Resources
  customDbUrl:
  eissn: 2732-5121
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0002513619
  issn: 2732-5121
  databaseCode: M~E
  dateStart: 20210101
  isFulltext: true
  titleUrlDefault: https://road.issn.org
  providerName: ISSN International Centre
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwrV3BbtQwELWg4sAFtQLEAq2mEgc4hCa2k9jl1FYtHKDqoaCVELJie1z2kpbd7h75F_6UGWdbbU5cuOSQOInlmdjvTTxvhHiTvJayRFUoVKHQEpvCW5UK00WbUpLRZsWbb5_b83MzndqLjVJfvCdskAceBu4gdFHKoDurmqSRy5NETEYjeiJv3gaefcvWbpApnoNp1VZEDdYpwbIhmsfVqDijJwe5OahSqvfVaDXKov0jpDneJ7mx8JxtiydrxAhHQ093xAPsn4o_G_oY_RXQ8Mx-LbM4KtAzWBaCQ-DAUVa4ysrSPK0B7_nIRSEwcrsBsuIc3n68uHxXKMCcSRXhPtIGWdB6cQhHEGi9g6xGC99XQ5QNqg9wg3T_kABzCBKyRvkK449n4uvZ6eXJp2JdbaEIilBAgWhMCq3stG0sEi4LdekrmTRhiJCi8kgsOjaNjrWvkk1SN3VKhhBAUC36oJ6Lrf66xxcCpCISZcqkjO90LcuuKdtExJCgXES09UQc3I26uxlENRyTEbaTG9nJZTu5aiKO2Tj3rVkUO58gV3FrV3H_cpWJ2L8zraOPiP-MdD1eLxdOmpp4p66MnQgzsvnojeMr_exnluMmQqtrU-qX_6OPr8RjLmg_BHlei63b-RJ3xaOwup0t5nviYTs1e9nV6fjl9-lfwR8MxA
linkProvider Directory of Open Access Journals
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=Investigating+antiquities+trafficking+with+generative+pre-trained+transformer+%28GPT%29-3+enabled+knowledge+graphs%3A+A+case+study+%5Bversion+1%3B+peer+review%3A+2+approved%5D&rft.jtitle=Open+research+Europe&rft.au=Shawn+Graham&rft.au=Donna+Yates&rft.au=Ahmed+El-Roby&rft.date=2023&rft.pub=F1000+Research+Ltd&rft.eissn=2732-5121&rft.volume=3&rft_id=info:doi/10.12688%2Fopenreseurope.16003.1&rft.externalDBID=DOA&rft.externalDocID=oai_doaj_org_article_cad22c4a936f4e6184def84eeb788b9c
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=2732-5121&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=2732-5121&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=2732-5121&client=summon