Building a Semantic Knowledge Graph - web.pdf

Presentation to European Bibframe Workshop 2024 - Helsinki.
Description of the National Library Board Singapore semantic knowledge graph and its developments.

1. A Semantic Knowledge Graph at
National Library Board Singapore
BIBFRAME Workshop in Europe 2024
17
th
September 2024 -Helsinki
Richard Wallis
Evangelist and Founder
Data Liberate
[email protected]

2. Independent Consultant, Evangelist & Founder
W3C Community Groups:
•Bibframe2Schema (Chair) –Standardised conversion path(s)
•Schema Bib Extend (Chair) - Bibliographic data
•Schema Architypes (Chair) - Archives
•Financial Industry Business Ontology –Financial schema.org
•Tourism Structured Web Data (Co-Chair)
•Schema Course Extension
•Schema IoT Community
•Educational & Occupational Credentials in Schema.org
[email protected] — @dataliberate
40+ Years – Computing
30+ Years – Cultural Heritage technology
20+ Years – Semantic Web & Linked Data
Worked With:
•Google – Schema.org vocabulary, site, extensions. documentation and community
•OCLC – Global library cooperative
•FIBO – Financial Industry Business Ontology Group
•Various Clients – Implementing/understanding Linked Data, Schema.org:
National Library Board Singapore
British Library — Stanford University — Europeana
2

3. 3
Agenda for today

4. 4
Agenda for today
•National Library and their resources

5. 5
Agenda for today
•National Library and their resources
•Knowledge Graph ambition

6. 6
Agenda for today
•National Library and their resources
•Knowledge Graph ambition
•Linked Data Management System –the LDMS delivered

7. 7
Agenda for today
•National Library and their resources
•Knowledge Graph ambition
•Linked Data Management System –the LDMS delivered
•Continued development

8. 8
Agenda for today
•National Library and their resources
•Knowledge Graph ambition
•Linked Data Management System –the LDMS delivered
•Continued development
–Data sharing with the Entity Data Service

9. 9
Agenda for today
•National Library and their resources
•Knowledge Graph ambition
•Linked Data Management System –the LDMS delivered
•Continued development
–Data sharing with the Entity Data Service
–User experience enrichment –the sidebar API

10. 10
Agenda for today
•National Library and their resources
•Knowledge Graph ambition
•Linked Data Management System –the LDMS delivered
•Continued development
–Data sharing with the Entity Data Service
–User experience enrichment –the sidebar API
–Data quality enhancement utilizing external authorities

11. 11
National Library Board Singapore
PublicLibraries
Networkof28PublicLibraries,
including2 partnerlibraries*
ReadingProgrammesandInitiatives
ProgrammesandExhibitions
targetedat Singaporecommunities
*Partnerlibrariesarelibrarieswhicharepartnerowned
and
fundedbutmanagedbyNLB/NLB’ssubsidiaryLibraries
and ArchivesSolutionsPteLtd.Library@Chinatownand
theLifelong LearningInstituteLibraryarePartnerlibraries.
NationalArchives
TransferredfromNHBtoNLBinNov
2012
CustodianofSingapore’s
Collective Memory:Responsible
forCollection, Preservationand
Managementof
Singapore’s Publicand Private
ArchivalRecords
PromotesPublicInterestinour
Nation’s
HistoryandHeritage
NationalLibrary
PreservingSingapore’sPrint
and LiteraryHeritage,and
Intellectual memory
ReferenceCollections
LegalDeposit(including
electronic)

12. 12
Over
560,000
Singapore&
SEAitems
Over147,000
Chinese,Malay &
TamilLanguages
items
ReferenceCollection
Over62,000
SocialSciences
& Humanities
items
Over39,000
Science &
Technology
items
Over
53,000
Artsitems
Over19,000
RareMaterials
items
ArchivalMaterials
Over290,000
Governmentfiles&
Parliamentpapers
Over190,000
Audiovisual&sound
recordings
Over70,000
Maps & building
plans
Over
1.14m
Photographs
Over35,000
Oral history
interviews
Over55,000
Speeches&press
releases
Over
7,000
Posters
National Library Board Singapore
Over5m
printcollection
Over2.4m
musictracks
78
databases
Over7,400
e-newspapersand
e-magazinestitles
Over
8,000
e-learning
courses
Over1.7m
e-booksand
audiobooks
LendingCollection

13. 13
National Library Board Online Services

14. 15
The Ambition
•To enable the discovery & display of entitles from different sources
in a combined interface

15. 16
The Ambition
•To enable the discovery & display of entitles from different sources
in a combined interface
•To bring together resources physical and digital

16. 17
The Ambition
•To enable the discovery & display of entitles from different sources
in a combined interface
•To bring together resources physical and digital
•To bring together diverse systems across the National Library,
National Archives, and Public Libraries in a Linked Data Environment

17. 18
The Ambition
•To enable the discovery & display of entitles from different sources
in a combined interface
•To bring together resources physical and digital
•To bring together diverse systems across the National Library,
National Archives, and Public Libraries in a Linked Data Environment
•To provide a staff interface to view and manage all entities, their
descriptions and relationships

18. 19
The Ambition –Technical Challenges
•To produce a Knowledge Graph that is [daily] up to date

19. 20
The Ambition –Technical Challenges
•To produce a Knowledge Graph that is [daily] up to date
•Not to replace current cataloging processes & practices

20. 21
The Ambition –Technical Challenges
•To produce a Knowledge Graph that is [daily] up to date
•Not to replace current cataloging processes & practices
–Marc cataloguing in the ILS

21. 22
The Ambition –Technical Challenges
•To produce a Knowledge Graph that is [daily] up to date
•Not to replace current cataloging processes & practices
–Marc cataloguing in the ILS
–TTE maintenance in authority control

22. 23
The Ambition –Technical Challenges
•To produce a Knowledge Graph that is [daily] up to date
•Not to replace current cataloging processes & practices
–Marc cataloguing in the ILS
–TTE maintenance in authority control
–Dublin Core content management for CMS sites and Archives

23. 24
The Ambition –Technical Challenges
•To produce a Knowledge Graph that is [daily] up to date
•Not to replace current cataloging processes & practices
–Marc cataloguing in the ILS
–TTE maintenance in authority control
–Dublin Core content management for CMS sites and Archives
•Data sharable with the world
–Linked Open Data
–Schema.org

24. 25
The Ambition –Technical Challenges
•To produce a Knowledge Graph that is [daily] up to date
•Not to replace current cataloging processes & practices
–Marc cataloguing in the ILS
–TTE maintenance in authority control
–Dublin Core content management for CMS sites and Archives
•Data sharable with the world
–Linked Open Data
–Schema.org
•An aggregated source of truth

25. 26
Contract Awarded
metaphactoryplatform
Low-code knowledge graph platform
Semantic knowledge modeling
Semantic search & discovery
AWS Partner
Public sector partner
Singapore based
Linked Data, Structured data, Semantic
Web, bibliographic meta data, Schema.org
and management systems consultant

26. 27
Basic Data Model
•Linked Data
–BIBFRAME to capture detail of bibliographic records
–Schema.org to deliver structured data for search engines

27. 28
Basic Data Model
•Linked Data
–BIBFRAME to capture detail of bibliographic records
–Schema.org to deliver structured data for search engines
–Schema.org representation of CMS, NAS, TTE data

28. 29
Basic Data Model
•Linked Data
–BIBFRAME to capture detail of bibliographic records
–Schema.org to deliver structured data for search engines
–Schema.org representation of CMS, NAS, TTE data
–Schema.org enrichment of BIBFRAME

29. 30
Basic Data Model
•Linked Data
–BIBFRAME to capture detail of bibliographic records
–Schema.org to deliver structured data for search engines
–Schema.org representation of CMS, NAS, TTE data
–Schema.org enrichment of BIBFRAME
•Schema.org as the ‘lingua franca’ vocabulary of the Knowledge graph

30. 31
Basic Data Model
•Linked Data
–BIBFRAME to capture detail of bibliographic records
–Schema.org to deliver structured data for search engines
–Schema.org representation of CMS, NAS, TTE data
–Schema.org enrichment of BIBFRAME
•Schema.org as the ‘lingua franca’ vocabulary of the Knowledge graph
–All entities described using Schema.org as a minimum.

31. 32
Data Data Data!
Data Source Source Records Entity CountUpdate Frequency
ILS 1.4m 7.9m Daily
CMS 82k 228k Weekly
NAS 1.6m 6.7m Monthly
TTE 3k 317k Monthly
3.1m 15.15m

32. 33
Data Ingest Pipelines
•Triggered by data upload from source system

33. 34
Data Ingest Pipelines
•Triggered by data upload from source system
•ILS –daily

34. 35
Data Ingest Pipelines
•Triggered by data upload from source system
•ILS –daily
–MARC-XML parsed through Open Source scripts:
•Marc2bibframe2 –Library of Congress
•Bibframe2schema –Bibframe2Schema.org

35. 36
Data Ingest Pipelines
•Triggered by data upload from source system
•ILS –daily
–MARC-XML parsed through Open Source scripts:
•Marc2bibframe2 –Library of Congress
•Bibframe2schema –Bibframe2Schema.org
•TTE Authorities –Monthly

36. 37
Data Ingest Pipelines
•Triggered by data upload from source system
•ILS –daily
–MARC-XML parsed through Open Source scripts:
•Marc2bibframe2 –Library of Congress
•Bibframe2schema –Bibframe2Schema.org
•TTE Authorities –Monthly
–Bespoke CSV conversion

37. 38
Data Ingest Pipelines
•Triggered by data upload from source system
•ILS –daily
–MARC-XML parsed through Open Source scripts:
•Marc2bibframe2 –Library of Congress
•Bibframe2schema –Bibframe2Schema.org
•TTE Authorities –Monthly
–Bespoke CSV conversion
•CMS & NAS –Weekly / Monthly

38. 39
Data Ingest Pipelines
•Triggered by data upload from source system
•ILS –daily
–MARC-XML parsed through Open Source scripts:
•Marc2bibframe2 –Library of Congress
•Bibframe2schema –Bibframe2Schema.org
•TTE Authorities –Monthly
–Bespoke CSV conversion
•CMS & NAS –Weekly / Monthly
–Dublin Core to Schema.org

39. 40
Technical Architecture (simplified)
Hosted on Amazon Web Services
Batch Scripts
import control
Etc.
SOURCE DATA
IMPORT

40. 41
Technical Architecture (simplified)
Hosted on Amazon Web Services
Pipeline
processing
Batch Scripts
import control
Etc.
SOURCE DATA
IMPORT

41. 42
Technical Architecture (simplified)
Hosted on Amazon Web Services
GraphDB
Cluster
GraphDB
Cluster
GraphDB
Cluster
GraphDB
Cluster
Pipeline
processing
Batch Scripts
import control
Etc.
SOURCE DATA
IMPORT

42. 43
Technical Architecture (simplified)
Hosted on Amazon Web Services
EDS
GraphDB
Cluster
GraphDB
Cluster
GraphDB
Cluster
GraphDB
Cluster
Pipeline
processing
Batch Scripts
import control
Etc.
SOURCE DATA
IMPORT

43. 44
Technical Architecture (simplified)
Hosted on Amazon Web Services
EDS
GraphDB
Cluster
GraphDB
Cluster
GraphDB
Cluster
GraphDB
Cluster
Pipeline
processing
Batch Scripts
import control
Etc.
SOURCE DATA
IMPORT
DMI

44. 45
A need for entity reconciliation …..
•Lots (and lots and lots) of source entities –10 million entities

45. 46
A need for entity reconciliation …..
•Lots (and lots and lots) of source entities –10 million entities
•Lots of duplication
–Lee, KuanYew –1
st
Prime Minister of Singapore
•160 individual entities in ILS source data

46. 47
A need for entity reconciliation …..
•Lots (and lots and lots) of source entities –10 million entities
•Lots of duplication
–Lee, KuanYew –1
st
Prime Minister of Singapore
•160 individual entities in ILS source data
–Singapore Art Museum
•Entities from source data
•21 CMS, 1 NAS, 66 ILS, 1 TTE

47. 48
A need for entity reconciliation …..
•Lots (and lots and lots) of source entities –10 million entities
•Lots of duplication
–Lee, KuanYew –1
st
Prime Minister of Singapore
•160 individual entities in ILS source data
–Singapore Art Museum
•Entities from source data
•21 CMS, 1 NAS, 66 ILS, 1 TTE
•Users only want 1 of each!

48. 49
Adaptive Data Model Concepts
•Source entitles
–Individual representation of source data

49. 50
Adaptive Data Model Concepts
•Source entitles
–Individual representation of source data
•Aggregation entities
–Tracking relationships between source entities for the same thing
–No copying of attributes

50. 51
Adaptive Data Model Concepts
•Source entitles
–Individual representation of source data
•Aggregation entities
–Tracking relationships between source entities for the same thing
–No copying of attributes
•Primary Entities
–Searchable by users
–Displayable to users
–Consolidation of aggregated source data & managed attributes

51. 52

52. 53

53. 54

54. 55

55. 56

56. 57

57. 58

58. 59

59. 60

60. 66
The entity iceberg

61. 67
The entity iceberg
Primary

62. 68
The entity iceberg
Primary
Discovery

63. 69
The entity iceberg
Primary
Aggregation
Discovery

64. 70
The entity iceberg
Primary
Aggregation
Source
Ingestion
Pipelines
Discovery

65. 71
The entity iceberg
Primary
Aggregation
Source
Ingestion
Pipelines
Discovery
Management

66. 72
The NLB Knowledge Graph
•666M Triples

67. 73
The NLB Knowledge Graph
•666M Triples
•10M Source Entities

68. 74
The NLB Knowledge Graph
•666M Triples
•10M Source Entities
•5.8M Primary Entities

69. 75
The NLB Knowledge Graph
•666M Triples
•10M Source Entities
•5.8M Primary Entities
–Aggregation of source derived entities

70. 76
The NLB Knowledge Graph
•666M Triples
•10M Source Entities
•5.8M Primary Entities
–Aggregation of source derived entities
–Searchable

71. 77
The NLB Knowledge Graph
•666M Triples
•10M Source Entities
•5.8M Primary Entities
–Aggregation of source derived entities
–Searchable
–Shared with world

72. 78
The NLB Knowledge Graph
•666M Triples
•10M Source Entities
•5.8M Primary Entities
–Aggregation of source derived entities
–Searchable
–Shared with world

73. 79
NLB Linked Data Management System (LDMS)
•Powered by the Knowledge Graph

74. 80
NLB Linked Data Management System (LDMS)
•Powered by the Knowledge Graph
•Updated daily

75. 81
NLB Linked Data Management System (LDMS)
•Powered by the Knowledge Graph
•Updated daily
•A new separate environment built on established systems

76. 82
NLB Linked Data Management System (LDMS)
•Powered by the Knowledge Graph
•Updated daily
•A new separate environment built on established systems
•No changes in cataloguing practices

77. 83
NLB Linked Data Management System (LDMS)
•Powered by the Knowledge Graph
•Updated daily
•A new separate environment built on established systems
•No changes in cataloguing practices
•No cataloguer retraining

78. 84
NLB Linked Data Management System (LDMS)
•Powered by the Knowledge Graph
•Updated daily
•A new separate environment built on established systems
•No changes in cataloguing practices
•No cataloguer retraining
•Not just the bibliographic (MARC) data

79. 85
NLB Linked Data Management System (LDMS)
•Powered by the Knowledge Graph
•Updated daily
•A new separate environment built on established systems
•No changes in cataloguing practices
•No cataloguer retraining
•Not just the bibliographic (MARC) data
•No replacement systems –to implement Linked Data

80. 86
NLB Linked Data Management System (LDMS)
•Powered by the Knowledge Graph
•Updated daily
•A new separate environment built on established systems
•No changes in cataloguing practices
•No cataloguer retraining
•Not just the bibliographic (MARC) data
•No replacement systems –to implement Linked Data
–MARC based ILS swap out occurred mid project –without LDMS impact

81. 87
NLB Linked Data Management System (LDMS)
•Powered by the Knowledge Graph
•Updated daily
•A new separate environment built on established systems
•No changes in cataloguing practices
•No cataloguer retraining
•Not just the bibliographic (MARC) data
•No replacement systems –to implement Linked Data
–MARC based ILS swap out occurred mid project –without LDMS impact
•Delivering Linked Data benefits back into the organization

82. 88
Building on the Knowledge Graph
Entity Data Service
•Open Linked Data interface

83. 89
Building on the Knowledge Graph
Entity Data Service
•Open Linked Data interface
•Dereferencing entity URIs

84. 90
Building on the Knowledge Graph
Entity Data Service
•Open Linked Data interface
•Dereferencing entity URIs
•Content negotiation for RDF/XML / JSON-LD / Turtle / N-Triples

85. 91
Building on the Knowledge Graph
Entity Data Service
•Open Linked Data interface
•Dereferencing entity URIs
•Content negotiation for RDF/XML / JSON-LD / Turtle / N-Triples
•Download formats RDF/XML / JSON-LD / Turtle / N-Triples

86. 92
Building on the Knowledge Graph
Entity Data Service
•Open Linked Data interface
•Dereferencing entity URIs
•Content negotiation for RDF/XML / JSON-LD / Turtle / N-Triples
•Download formats RDF/XML / JSON-LD / Turtle / N-Triples
•Embedded Schema.org

87. 93
Building on the Knowledge Graph
Entity Data Service
•Open Linked Data interface
•Dereferencing entity URIs
•Content negotiation for RDF/XML / JSON-LD / Turtle / N-Triples
•Download formats RDF/XML / JSON-LD / Turtle / N-Triples
•Embedded Schema.org
•Enhanced navigation

91. 97
Building on the Knowledge Graph
Enriching the User Journey

92. 98
Building on the Knowledge Graph
Enriching the User Journey
•Systems are often silos

93. 99
Building on the Knowledge Graph
Enriching the User Journey
•Systems are often silos
•User search and navigation constrained to the data in the silo

94. 100
Building on the Knowledge Graph
Enriching the User Journey
•Systems are often silos
•User search and navigation constrained to the data in the silo
•Knowledge Graph populated from several individual systems

95. 101
Building on the Knowledge Graph
Enriching the User Journey
•Systems are often silos
•User search and navigation constrained to the data in the silo
•Knowledge Graph populated from several individual systems
•Entities aggregated and related across system sources

96. 102
Building on the Knowledge Graph
Enriching the User Journey
•Systems are often silos
•User search and navigation constrained to the data in the silo
•Knowledge Graph populated from several individual systems
•Entities aggregated and related across system sources
•The fuel to explore between systems

97. 103
Building on the Knowledge Graph
Enriching the User Journey
•Systems are often silos
•User search and navigation constrained to the data in the silo
•Knowledge Graph populated from several individual systems
•Entities aggregated and related across system sources
•The fuel to explore between systems
•Via a navigational interface sidebar –bridging the silos

98. 104
Building on the Knowledge Graph
Enriching the User Journey
•Systems are often silos
•User search and navigation constrained to the data in the silo
•Knowledge Graph populated from several individual systems
•Entities aggregated and related across system sources
•The fuel to explore between systems
•Via a navigational interface sidebar –bridging the silos
•Plugged into user interface

99. 105
Building on the Knowledge Graph
Enriching the User Journey
•Systems are often silos
•User search and navigation constrained to the data in the silo
•Knowledge Graph populated from several individual systems
•Entities aggregated and related across system sources
•The fuel to explore between systems
•Via a navigational interface sidebar –bridging the silos
•Plugged into user interface
•Powered by a JavaScript Sidebar API

102. Use of the JavaScript Sidebar API

103. Use of the JavaScript Sidebar API
→ API call to KG –article ID passed as parameter

104. Use of the JavaScript Sidebar API
→ API call to KG –article ID passed as parameter
← Description of associated Primary entity returned

105. Description includes list of ‘about’ related entity IDs
used to build display and navigation links
Use of the JavaScript Sidebar API
→ API call to KG –article ID passed as parameter
← Description of associated Primary entity returned

106. Clicking sidebar links trigger new API calls to rebuild
the sidebar display as entity relationships are
followed
Description includes list of ‘about’ related entity IDs
used to build display and navigation links
Use of the JavaScript Sidebar API
→ API call to KG –article ID passed as parameter
← Description of associated Primary entity returned

107. Clicking sidebar links trigger new API calls to rebuild
the sidebar display as entity relationships are
followed
Knowledge Graph navigation via a sidebar
Description includes list of ‘about’ related entity IDs
used to build display and navigation links
Use of the JavaScript Sidebar API
→ API call to KG –article ID passed as parameter
← Description of associated Primary entity returned

108. Clicking sidebar links trigger new API calls to rebuild
the sidebar display as entity relationships are
followed
Knowledge Graph navigation via a sidebar
Description includes list of ‘about’ related entity IDs
used to build display and navigation links
Use of the JavaScript Sidebar API
→ API call to KG –article ID passed as parameter
← Description of associated Primary entity returned

109. 115
KG Quality Enhancement from Authorities
LCNAF URI Ingestion
•For Person / Organization entities with LCNAF URIs

110. 116
KG Quality Enhancement from Authorities
LCNAF URI Ingestion
•For Person / Organization entities with LCNAF URIs
•Created via the marc2bibframe2 scripts -from $0subfield

111. 117
KG Quality Enhancement from Authorities
LCNAF URI Ingestion
•For Person / Organization entities with LCNAF URIs
•Created via the marc2bibframe2 scripts -from $0subfield
•Create rdfs:labelvalues from the marc record
eg.700$a+ 700$d

112. 118
KG Quality Enhancement from Authorities
LCNAF URI Ingestion
•For Person / Organization entities with LCNAF URIs
•Created via the marc2bibframe2 scripts -from $0subfield
•Create rdfs:labelvalues from the marc record
eg.700$a+ 700$d
•These values are not controlled –entity can have several different labels

113. 119
KG Quality Enhancement from Authorities
LCNAF URI Ingestion
•For Person / Organization entities with LCNAF URIs
•Created via the marc2bibframe2 scripts -from $0subfield
•Create rdfs:labelvalues from the marc record
eg.700$a+ 700$d
•These values are not controlled –entity can have several different labels
•Use LCNAF authority data to introduce naming consistency

114. 120
KG Quality Enhancement from Authorities
LCNAF URI Ingestion
•For Person / Organization entities with LCNAF URIs
•Created via the marc2bibframe2 scripts -from $0subfield
•Create rdfs:labelvalues from the marc record
eg.700$a+ 700$d
•These values are not controlled –entity can have several different labels
•Use LCNAF authority data to introduce naming consistency
•Lookup against LCNAF to identify & ingest authoritative version

115. 121
KG Quality Enhancement from Authorities
LCNAF URI Ingestion
•For Person / Organization entities with LCNAF URIs
•Created via the marc2bibframe2 scripts -from $0subfield
•Create rdfs:labelvalues from the marc record
eg.700$a+ 700$d
•These values are not controlled –entity can have several different labels
•Use LCNAF authority data to introduce naming consistency
•Lookup against LCNAF to identify & ingest authoritative version
•LCNAF values take precedence in primary entity consolidation

116. Quality Enrichment from Authorities
LCNAF URI Ingestion
MARC XML:

117. Quality Enrichment from Authorities
LCNAF URI Ingestion
MARC XML:

118. Quality Enrichment from Authorities
LCNAF URI Ingestion
MARC XML:
Bibframe RDF:

119. Quality Enrichment from Authorities
LCNAF URI Ingestion
MARC XML:
Bibframe RDF:

120. Quality Enrichment from Authorities
LCNAF URI Ingestion
MARC XML:
Bibframe RDF:
MARC XML:

121. Quality Enrichment from Authorities
LCNAF URI Ingestion
MARC XML:
Bibframe RDF:
MARC XML:
Bibframe RDF:

122. Quality Enrichment from Authorities
LCNAF URI Ingestion
MARC XML:
Bibframe RDF:
MARC XML:
Bibframe RDF:
Entity result in Knowledge Graph
Which is correct?

123. Quality Enrichment from Authorities
LCNAF URI Ingestion
MARC XML:
Bibframe RDF:
MARC XML:
Bibframe RDF:
Entity result in Knowledge Graph
Which is correct?

124. Quality Enrichment from Authorities
LCNAF URI Ingestion
MARC XML:
Bibframe RDF:
MARC XML:
Bibframe RDF:
Entity result in Knowledge Graph
Which is correct?
Ingest from LCNAF and give precedence in consolidation

125. 131
Quality Enhancement from Authorities
LCNAF Person & Organization Name Matching
•For all Person and Organization primary entities

126. 132
Quality Enhancement from Authorities
LCNAF Person & Organization Name Matching
•For all Person and Organization primary entities
•Perform a string-matching LCNAF lookup for schema:namevalues

127. 133
Quality Enhancement from Authorities
LCNAF Person & Organization Name Matching
•For all Person and Organization primary entities
•Perform a string-matching LCNAF lookup for schema:namevalues
•Automatic background process

128. 134
Quality Enhancement from Authorities
LCNAF Person & Organization Name Matching
•For all Person and Organization primary entities
•Perform a string-matching LCNAF lookup for schema:namevalues
•Automatic background process
•If exact match
–Ingest LCNAF entity –takes precedence in consolidation

129. 135
Quality Enhancement from Authorities
LCNAF Person & Organization Name Matching
•For all Person and Organization primary entities
•Perform a string-matching LCNAF lookup for schema:namevalues
•Automatic background process
•If exact match
–Ingest LCNAF entity –takes precedence in consolidation
•If close match
–Add to list of match candidates

130. 136
Quality Enhancement from Authorities
LCNAF Person & Organization Name Matching
•For all Person and Organization primary entities
•Perform a string-matching LCNAF lookup for schema:namevalues
•Automatic background process
•If exact match
–Ingest LCNAF entity –takes precedence in consolidation
•If close match
–Add to list of match candidates
–[Human] curator either accepts as a match or not

131. Quality Enrichment from Authorities
LCNAF Person & Organization Name Matching

132. Quality Enrichment from Authorities
LCNAF Person & Organization Name Matching

133. 139
•2 years in development
NLB Linked Data Management System (LDMS)

134. 140
•2 years in development
•Live and operational for 1.5 years
NLB Linked Data Management System (LDMS)

135. 141
•2 years in development
•Live and operational for 1.5 years
•Built on a 666M triple Knowledge Graph
NLB Linked Data Management System (LDMS)

136. 142
•2 years in development
•Live and operational for 1.5 years
•Built on a 666M triple Knowledge Graph
•Automatically updated daily
NLB Linked Data Management System (LDMS)

137. 143
•2 years in development
•Live and operational for 1.5 years
•Built on a 666M triple Knowledge Graph
•Automatically updated daily
•Using Bibframe & Schema.org
NLB Linked Data Management System (LDMS)

138. 144
•2 years in development
•Live and operational for 1.5 years
•Built on a 666M triple Knowledge Graph
•Automatically updated daily
•Using Bibframe & Schema.org
•Built on –not replacing –established systems & practices
NLB Linked Data Management System (LDMS)

139. 145
•2 years in development
•Live and operational for 1.5 years
•Built on a 666M triple Knowledge Graph
•Automatically updated daily
•Using Bibframe & Schema.org
•Built on –not replacing –established systems & practices
•A Linked Data Service for NLB
NLB Linked Data Management System (LDMS)

140. 146
•2 years in development
•Live and operational for 1.5 years
•Built on a 666M triple Knowledge Graph
•Automatically updated daily
•Using Bibframe & Schema.org
•Built on –not replacing –established systems & practices
•A Linked Data Service for NLB
–Utilizing external authorities to enrich and standardize descriptions
NLB Linked Data Management System (LDMS)

141. 147
•2 years in development
•Live and operational for 1.5 years
•Built on a 666M triple Knowledge Graph
•Automatically updated daily
•Using Bibframe & Schema.org
•Built on –not replacing –established systems & practices
•A Linked Data Service for NLB
–Utilizing external authorities to enrich and standardize descriptions
–Part of Open Linked Data Cloud –via Entity Data Service
NLB Linked Data Management System (LDMS)

142. 148
•2 years in development
•Live and operational for 1.5 years
•Built on a 666M triple Knowledge Graph
•Automatically updated daily
•Using Bibframe & Schema.org
•Built on –not replacing –established systems & practices
•A Linked Data Service for NLB
–Utilizing external authorities to enrich and standardize descriptions
–Part of Open Linked Data Cloud –via Entity Data Service
–Enriching user journeys on non-linked data systems –via sidebar API
NLB Linked Data Management System (LDMS)

143. A Semantic Knowledge Graph at
National Library Board Singapore
BIBFRAME Workshop in Europe 2024
17
th
September 2024 -Helsinki
Richard Wallis
Evangelist and Founder
Data Liberate
[email protected]