Andreas Blumauer
CEO, Semantic Web Company
Leveraging Knowledge
Graphs in your
Enterprise Knowledge
Management System
Use PoolParty 7.0
to manage Knowledge
Graphs along the whole
Linked Data Life Cycle
CMS/DMS
/DAM/..
Graph-based
Introduction
2
Semantic Web
Company
Founder &
CEO of
Andreas
Blumauer
developer &
vendor of
2004founded
7.0version
active at
based on
headquartered
part of
Knowledge
Graphs
manages
standard for
part of
>200
serves customers
Taxonomies
Ontologies
standard for
graduates
Text
Mining
used for
Graph
database
integrates
with
PoolParty
Software Ltd
Director of
parent
company of
London
located
named
by
Vienna
Gartner
KMWorld
Search
engine
Current Status of
the Graph Market
Moving towards Semantic AI
4Knowledge
Graph Adoption
Source: Collapsing the IT Stack: Clearing a path for AI adoption
Alan Morrison (Sr. Research Fellow at PwC)
5Hype Cycle for
Artificial
Intelligence,
2018
“Once structured in the form of a knowledge graph,
unstructured data can be queried, thereby preprocessing it for analysis.”
6Semantic AI
Fusing Machine
Learning with
Knowledge
Graphs
“...the use of graphs as a means to
better generalize from one instance of
a problem to another 1)
.”
1) Relational inductive biases, deep learning, and graph networks
2) AI Requires More Than Machine Learning (via Forbes)
3) DARPA Embraces ‘Common Sense’ Approach to AI
“The confluence of symbolic
reasoning and machine learning
enables the enterprise to solve an
assortment of complicated business
problems applicable to real-world
situations -- as opposed to simply
automating facile, repetitive tasks. 2)
.”
“The absence of common sense
prevents an intelligent system from …
communicating naturally with people.
3)
.”
7The fast growing
Graph Database
Market
Amazon Neptune Azure Cosmos DB
▸ Stardog
▸ Marklogic
▸ AllegroGraph
▸ GraphDB
▸ Oracle Spatial&Graph
▸ Neo4j
▸ ...
Property Graph RDF Graph (Triple Stores)
Main use case Traverse a graph Query a graph
Typical
applications
Path Analytics,
Social Network Analysis
Data Integration,
Knowledge Representation
Standards No standards
→ Gremlin, Cypher, PGQL, ...
W3C Semantic Web standards
→ SPARQL 1.1
Additional
options
Shortest path calculations Inferencing
Core Principles
Things - not Strings, Semantic Layer,
Linked Data Life Cycle
“Things but not Strings”: Semantic Knowledge
Graphs manage resources, not just terms
http://www.my.com/
taxonomy/62346723
prefLabel
Retina
image
http://www.my.com/
images/90546089
http://www.my.com/
taxonomy/
97345854
prefLabel
Funduscope
altLabel
Ophthalmoscope
http://www.mycom.com
/taxonomy/4543567
prefLabel
Diagnostic Equipment
has broader
Core Principle
The Semantic
Layer completes
the Four-layered
Data & Content
Architecture
10
(= Enterprise KG)
Knowledge
Graphs as input
for Machine
Learning
11 Unstructured Data
Structured Data
Other Domain-
Specific Graphs
Machine
Learning
Enterprise
Knowledge
Graph
Cognitive
Applications
Use Cases
for (Enterprise)
Knowledge Graphs
Semantic Layer,
Linked Data Life Cycle
13Five Generic
Use Cases for
Graphs
1. dealing with hierarchical or highly connected datasets
2. entity-centric views (in contrast to document-centric views)
3. exploring the connections between the entities of a graph
4. integrating heterogeneous data sources
(structured & unstructured, “schema-late” approach)
5. federated (unified) views across multiple data silos within the
enterprise
Use Cases
across various
industries
14
15Example:
Citizen portal
healthdirect.gov.au/
As a citizen I want to receive guidance to
find reliable health information,
including
● articles from trusted sources
● information about drugs and
medicines
● medical services
● guidance along symptoms
Trusted health
information
Australian Health
Thesaurus
DrugBank
→ Linking Structured Data and Documents
to Industry Knowledge Graphs
Australian Register of
Therapeutic Goods
16Example:
HR Analytics
As an HR manager, for upcoming
training programmes, I want to
identify employees who
● have a certain skill set
● have a specific degree
● have skills that are increasingly
important on the labour market
● fall into a specific salary range
Employee database
Resumes
Labour market statistics
→ Linking Structured to Unstructured Data
How it works
17
Employee
database
Resumes
Labour market
statistics
PoolParty UnifiedViews
RDF
Graph Database
PoolParty GraphSearch
PoolParty
Thesaurus Server
PoolParty
User
Now I can
identify
employees
along many
dimensions.
18Example:
Research in
Life Sciences
As a researcher in pharmaceutical
industry, I want to plan new
experiments more efficiently.
I want to know what’s already
available. I’m interested in former
experiments where
● certain genes were tested
● under specific treatment conditions
● in a target therapeutic area
● with help from categorisation
systems like ‘disease hierarchies’
UniProt, ChEMBL
Experiments
Documentation
MeSH
DrugBank
→ Linking Structured to Unstructured Data
and to Industry Knowledge Graphs
19Making Use of
Knowledge
Graphs
→ Knowledge Graphs serve as means to enrich unstructured information
to provide a rich set of additional access points to document repositories
Experiments
Document
Store
20Making Use of
Automated
Reasoning based
on Knowledge
Graphs
How to build a
Knowledge Graph?
Linked Data Life Cycle &
Anatomy of an Enterprise Knowledge Graph
Things and URIs
Venice
Peggy
Guggenheim
Museum
St. Mark’s
Square
http://my.com/1
http://my.com/2
http://my.com/3
Labels and basic relations:
Taxonomies and Thesauri
prefLabel
Venice
prefLabel
St. Mark’s Square
altLabel
Piazza
San Marco
Peggy
Guggenheim
Museum
prefLabel
Piazza
altLabel
Town Square
related
related
prefLabel
broader
http://my.com/1
http://my.com/2
http://my.com/3
http://my.com/4
Classes, specific relations, restrictions:
Ontologies and Custom Schemas
prefLabel
Venice
prefLabel
St. Mark’s Square
altLabel
Piazza
San Marco
http://schema.org/City
http://schema.org/TouristAttraction
http://schema.org/ArtGallery
Monday through
Sunday, all day
opening
Hours
image
http://schema.org/containedInPlace
prefLabel
Piazza
altLabel
Town Square
Peggy
Guggenheim
Museum
prefLabel
containedInPlace
containedInPlace
broader
Metadata and Graph annotations
prefLabel
Venice
prefLabel
St. Mark’s Square
altLabel
Piazza
San Marco
http://schema.org/City
http://schema.org/TouristAttraction
http://schema.org/ArtGallery
Monday through
Sunday, all day
opening
Hours
image
http://schema.org/containedInPlace
prefLabel
Piazza
altLabel
Town Square
Peggy
Guggenheim
Museum
prefLabel
containedInPlace
containedInPlace
CC BY-SA 3.0
broader
Entity linking and schema mappings:
Links to other graphs
prefLabel
Venice
prefLabel
St. Mark’s Square
altLabel
Piazza
San Marco
http://schema.org/City
http://schema.org/TouristAttraction
http://schema.org/ArtGallery
Monday through
Sunday, all day
opening
Hours
image
http://schema.org/containedInPlace
prefLabel
Piazza
altLabel
Town Square
Peggy
Guggenheim
Museum
prefLabel
CC BY-SA 3.0
broader
containedInPlace
containedInPlace
Linking to data and documents
stored in other systems
prefLabel
Venice
prefLabel
St. Mark’s Square
altLabel
Piazza
San Marco
http://schema.org/City
http://schema.org/TouristAttraction
http://schema.org/ArtGallery
Monday through
Sunday, all day
opening
Hours
image
http://schema.org/containedInPlace
prefLabel
Piazza
altLabel
Town Square
broader
Peggy
Guggenheim
Museum
prefLabel
containedInPlace
CC BY-SA 3.0
The Peggy
Guggenheim
Collection is
a modern art
museum on the
Grand Canal in
the Dorsoduro
sestiere of
Venice, Italy.
containedInPlace
Linked Data
Life Cycle
How to build
Enterprise
Knowledge
Graphs?
28
PoolParty
Platform
Functions and
Components
29
PoolParty
supports
Knowledge
Graph Building
along the whole
Life Cycle
30 ▸ UnifiedViews
▸ Extractor
▸ Thesaurus Server
▸ GraphEditor
▸ GraphEditor
▸ UnifiedViews
▸ UnifiedViews
▸ Extractor
▸ Thesaurus Server
▸ Extractor
▸ UnifiedViews
▸ Semantic Classifier
▸ UnifiedViews
▸ Thesaurus Server
▸ API
▸ GraphSearch
▸ 3rd
party
What’s new in
PoolParty 7.0?
Some Highlights and
new Features
PoolParty
GraphEditor
32
▸ create ontology-driven custom editors to work with graph data
▸ use multiple graphs to create integrated views on graph data
▸ import and export of RDF graphs
▸ benefit from assisted search over graph data
▸ benefit from assisted bulk editing of RDF graphs
▸ administrate graphs based on user-friendly inline editing
▸ generate SPARQL queries based on an assistant
PoolParty
Notifications
33
▸ stay informed on
changes in your
project
▸ configure multiple
notification settings
per project
▸ get notifications via
webhooks
▸ connect APIs to
consume notifications
Improved
Ontology
Management
34
▸ access your ontologies via a
tree view
▸ apply multilingual labels to
your classes, attributes and
relations
▸ define user group based
access rights on your
ontologies and custom
schemes
Improved User
Management
35
▸ access users/ roles/
groups via a tree
view
▸ an action-based role
management has
been implemented
▸ define
project-based roles
per user
Integration of
NER based on
Machine
Learning
▸ With 7.0, named entities can be extracted by using the
concept extract service (extract call).
▸ This complements PoolParty’s vocabulary-based
entity extraction method
▸ Two methods are now supported by default
▹ Maximum Entropy classification: person,
location, organisation (more specific classifiers
can be added programmatically)
▹ Rule-based recognition by using regex
expressions
36
PoolParty
UnifiedViews
Reworked and
improved GUI
37
PoolParty
UnifiedViews
New DPUs
▸ DBpedia batch linking
▸ RML mapping
▸ R2RML mapping
▸ JSON to XML transformation
▸ XML to JSON transformation
▸ ML-based extraction (max entropy)
▸ Thomson Reuters Open Calais extraction
▸ GraphSearch content preparator
▸ Ontology based tabular data mapping
▸ Data fusion
▸ Multi-threaded concept extraction
38
Visualisation
of Knowledge
Graphs
39
Visualisation
of Knowledge
Graphs
40
Get started
Start your own
Knowledge Graph project!
Maturity Model
Roadmap for a
more agile Data
Governance
Framework
42
Semantic Web
Starter Kit
43
Next steps
44 ▸ Webinar: PoolParty 7.0 (Technical)
▸ Semantic AI - White paper
▸ Test Account
▸ PoolParty Academy
CONNECT
Andreas Blumauer
CEO, Semantic Web Company GmbH
Director, PoolParty Software Ltd
▸ andreas.blumauer@semantic-web.com
▸ http://linkedin.com/in/andreasblumauer
▸ https://twitter.com/semwebcompany
45
© Semantic Web Company - http://www.semantic-web.at/ and http://www.poolparty.biz/

Leveraging Knowledge Graphs in your Enterprise Knowledge Management System

  • 1.
    Andreas Blumauer CEO, SemanticWeb Company Leveraging Knowledge Graphs in your Enterprise Knowledge Management System Use PoolParty 7.0 to manage Knowledge Graphs along the whole Linked Data Life Cycle
  • 2.
    CMS/DMS /DAM/.. Graph-based Introduction 2 Semantic Web Company Founder & CEOof Andreas Blumauer developer & vendor of 2004founded 7.0version active at based on headquartered part of Knowledge Graphs manages standard for part of >200 serves customers Taxonomies Ontologies standard for graduates Text Mining used for Graph database integrates with PoolParty Software Ltd Director of parent company of London located named by Vienna Gartner KMWorld Search engine
  • 3.
    Current Status of theGraph Market Moving towards Semantic AI
  • 4.
    4Knowledge Graph Adoption Source: Collapsingthe IT Stack: Clearing a path for AI adoption Alan Morrison (Sr. Research Fellow at PwC)
  • 5.
    5Hype Cycle for Artificial Intelligence, 2018 “Oncestructured in the form of a knowledge graph, unstructured data can be queried, thereby preprocessing it for analysis.”
  • 6.
    6Semantic AI Fusing Machine Learningwith Knowledge Graphs “...the use of graphs as a means to better generalize from one instance of a problem to another 1) .” 1) Relational inductive biases, deep learning, and graph networks 2) AI Requires More Than Machine Learning (via Forbes) 3) DARPA Embraces ‘Common Sense’ Approach to AI “The confluence of symbolic reasoning and machine learning enables the enterprise to solve an assortment of complicated business problems applicable to real-world situations -- as opposed to simply automating facile, repetitive tasks. 2) .” “The absence of common sense prevents an intelligent system from … communicating naturally with people. 3) .”
  • 7.
    7The fast growing GraphDatabase Market Amazon Neptune Azure Cosmos DB ▸ Stardog ▸ Marklogic ▸ AllegroGraph ▸ GraphDB ▸ Oracle Spatial&Graph ▸ Neo4j ▸ ... Property Graph RDF Graph (Triple Stores) Main use case Traverse a graph Query a graph Typical applications Path Analytics, Social Network Analysis Data Integration, Knowledge Representation Standards No standards → Gremlin, Cypher, PGQL, ... W3C Semantic Web standards → SPARQL 1.1 Additional options Shortest path calculations Inferencing
  • 8.
    Core Principles Things -not Strings, Semantic Layer, Linked Data Life Cycle
  • 9.
    “Things but notStrings”: Semantic Knowledge Graphs manage resources, not just terms http://www.my.com/ taxonomy/62346723 prefLabel Retina image http://www.my.com/ images/90546089 http://www.my.com/ taxonomy/ 97345854 prefLabel Funduscope altLabel Ophthalmoscope http://www.mycom.com /taxonomy/4543567 prefLabel Diagnostic Equipment has broader
  • 10.
    Core Principle The Semantic Layercompletes the Four-layered Data & Content Architecture 10 (= Enterprise KG)
  • 11.
    Knowledge Graphs as input forMachine Learning 11 Unstructured Data Structured Data Other Domain- Specific Graphs Machine Learning Enterprise Knowledge Graph Cognitive Applications
  • 12.
    Use Cases for (Enterprise) KnowledgeGraphs Semantic Layer, Linked Data Life Cycle
  • 13.
    13Five Generic Use Casesfor Graphs 1. dealing with hierarchical or highly connected datasets 2. entity-centric views (in contrast to document-centric views) 3. exploring the connections between the entities of a graph 4. integrating heterogeneous data sources (structured & unstructured, “schema-late” approach) 5. federated (unified) views across multiple data silos within the enterprise
  • 14.
  • 15.
    15Example: Citizen portal healthdirect.gov.au/ As acitizen I want to receive guidance to find reliable health information, including ● articles from trusted sources ● information about drugs and medicines ● medical services ● guidance along symptoms Trusted health information Australian Health Thesaurus DrugBank → Linking Structured Data and Documents to Industry Knowledge Graphs Australian Register of Therapeutic Goods
  • 16.
    16Example: HR Analytics As anHR manager, for upcoming training programmes, I want to identify employees who ● have a certain skill set ● have a specific degree ● have skills that are increasingly important on the labour market ● fall into a specific salary range Employee database Resumes Labour market statistics → Linking Structured to Unstructured Data
  • 17.
    How it works 17 Employee database Resumes Labourmarket statistics PoolParty UnifiedViews RDF Graph Database PoolParty GraphSearch PoolParty Thesaurus Server PoolParty User Now I can identify employees along many dimensions.
  • 18.
    18Example: Research in Life Sciences Asa researcher in pharmaceutical industry, I want to plan new experiments more efficiently. I want to know what’s already available. I’m interested in former experiments where ● certain genes were tested ● under specific treatment conditions ● in a target therapeutic area ● with help from categorisation systems like ‘disease hierarchies’ UniProt, ChEMBL Experiments Documentation MeSH DrugBank → Linking Structured to Unstructured Data and to Industry Knowledge Graphs
  • 19.
    19Making Use of Knowledge Graphs →Knowledge Graphs serve as means to enrich unstructured information to provide a rich set of additional access points to document repositories Experiments Document Store
  • 20.
    20Making Use of Automated Reasoningbased on Knowledge Graphs
  • 21.
    How to builda Knowledge Graph? Linked Data Life Cycle & Anatomy of an Enterprise Knowledge Graph
  • 22.
    Things and URIs Venice Peggy Guggenheim Museum St.Mark’s Square http://my.com/1 http://my.com/2 http://my.com/3
  • 23.
    Labels and basicrelations: Taxonomies and Thesauri prefLabel Venice prefLabel St. Mark’s Square altLabel Piazza San Marco Peggy Guggenheim Museum prefLabel Piazza altLabel Town Square related related prefLabel broader http://my.com/1 http://my.com/2 http://my.com/3 http://my.com/4
  • 24.
    Classes, specific relations,restrictions: Ontologies and Custom Schemas prefLabel Venice prefLabel St. Mark’s Square altLabel Piazza San Marco http://schema.org/City http://schema.org/TouristAttraction http://schema.org/ArtGallery Monday through Sunday, all day opening Hours image http://schema.org/containedInPlace prefLabel Piazza altLabel Town Square Peggy Guggenheim Museum prefLabel containedInPlace containedInPlace broader
  • 25.
    Metadata and Graphannotations prefLabel Venice prefLabel St. Mark’s Square altLabel Piazza San Marco http://schema.org/City http://schema.org/TouristAttraction http://schema.org/ArtGallery Monday through Sunday, all day opening Hours image http://schema.org/containedInPlace prefLabel Piazza altLabel Town Square Peggy Guggenheim Museum prefLabel containedInPlace containedInPlace CC BY-SA 3.0 broader
  • 26.
    Entity linking andschema mappings: Links to other graphs prefLabel Venice prefLabel St. Mark’s Square altLabel Piazza San Marco http://schema.org/City http://schema.org/TouristAttraction http://schema.org/ArtGallery Monday through Sunday, all day opening Hours image http://schema.org/containedInPlace prefLabel Piazza altLabel Town Square Peggy Guggenheim Museum prefLabel CC BY-SA 3.0 broader containedInPlace containedInPlace
  • 27.
    Linking to dataand documents stored in other systems prefLabel Venice prefLabel St. Mark’s Square altLabel Piazza San Marco http://schema.org/City http://schema.org/TouristAttraction http://schema.org/ArtGallery Monday through Sunday, all day opening Hours image http://schema.org/containedInPlace prefLabel Piazza altLabel Town Square broader Peggy Guggenheim Museum prefLabel containedInPlace CC BY-SA 3.0 The Peggy Guggenheim Collection is a modern art museum on the Grand Canal in the Dorsoduro sestiere of Venice, Italy. containedInPlace
  • 28.
    Linked Data Life Cycle Howto build Enterprise Knowledge Graphs? 28
  • 29.
  • 30.
    PoolParty supports Knowledge Graph Building along thewhole Life Cycle 30 ▸ UnifiedViews ▸ Extractor ▸ Thesaurus Server ▸ GraphEditor ▸ GraphEditor ▸ UnifiedViews ▸ UnifiedViews ▸ Extractor ▸ Thesaurus Server ▸ Extractor ▸ UnifiedViews ▸ Semantic Classifier ▸ UnifiedViews ▸ Thesaurus Server ▸ API ▸ GraphSearch ▸ 3rd party
  • 31.
    What’s new in PoolParty7.0? Some Highlights and new Features
  • 32.
    PoolParty GraphEditor 32 ▸ create ontology-drivencustom editors to work with graph data ▸ use multiple graphs to create integrated views on graph data ▸ import and export of RDF graphs ▸ benefit from assisted search over graph data ▸ benefit from assisted bulk editing of RDF graphs ▸ administrate graphs based on user-friendly inline editing ▸ generate SPARQL queries based on an assistant
  • 33.
    PoolParty Notifications 33 ▸ stay informedon changes in your project ▸ configure multiple notification settings per project ▸ get notifications via webhooks ▸ connect APIs to consume notifications
  • 34.
    Improved Ontology Management 34 ▸ access yourontologies via a tree view ▸ apply multilingual labels to your classes, attributes and relations ▸ define user group based access rights on your ontologies and custom schemes
  • 35.
    Improved User Management 35 ▸ accessusers/ roles/ groups via a tree view ▸ an action-based role management has been implemented ▸ define project-based roles per user
  • 36.
    Integration of NER basedon Machine Learning ▸ With 7.0, named entities can be extracted by using the concept extract service (extract call). ▸ This complements PoolParty’s vocabulary-based entity extraction method ▸ Two methods are now supported by default ▹ Maximum Entropy classification: person, location, organisation (more specific classifiers can be added programmatically) ▹ Rule-based recognition by using regex expressions 36
  • 37.
  • 38.
    PoolParty UnifiedViews New DPUs ▸ DBpediabatch linking ▸ RML mapping ▸ R2RML mapping ▸ JSON to XML transformation ▸ XML to JSON transformation ▸ ML-based extraction (max entropy) ▸ Thomson Reuters Open Calais extraction ▸ GraphSearch content preparator ▸ Ontology based tabular data mapping ▸ Data fusion ▸ Multi-threaded concept extraction 38
  • 39.
  • 40.
  • 41.
    Get started Start yourown Knowledge Graph project!
  • 42.
    Maturity Model Roadmap fora more agile Data Governance Framework 42
  • 43.
  • 44.
    Next steps 44 ▸Webinar: PoolParty 7.0 (Technical) ▸ Semantic AI - White paper ▸ Test Account ▸ PoolParty Academy
  • 45.
    CONNECT Andreas Blumauer CEO, SemanticWeb Company GmbH Director, PoolParty Software Ltd ▸ andreas.blumauer@semantic-web.com ▸ http://linkedin.com/in/andreasblumauer ▸ https://twitter.com/semwebcompany 45 © Semantic Web Company - http://www.semantic-web.at/ and http://www.poolparty.biz/