SlideShare a Scribd company logo
1 of 33
Marius Hartmann
Fraud Detection with Graphs
24. september 2019
Main task of The Danish Business Authority
2Danish Business Authority
Business service and registration of companies
Business development and digital growth
Business regulation and supervision planning
and rural business
EU and international affairs
Virk: The joint public one-stop shop to the
Danish business world
3Danish Business Authority
27.000.000
visits at Virk on an
annual basis
96%
of companies in
Denmark know
Virk
4.000.000
filings on Virk
annually
92%
Instant case
handling
ML Lab
 9 person strong
 Physics, astro-physics, economics, computer science, fine art, social
science, 7 Phd’s (+1 on the way)
 2/7 gender balance, kids and no-kids
Erhvervsstyrelsen 4
What can we do with ML and Graph?
 Help and guide users to make fewer mistakes
 Improve and scale our control and supervision
 Provide recommendations and personalize our solutions
 Improve our policy development with
ML created insight
5Danish Business Authority
Management
Owner
?
Example of control:
Strengthened company control regarding VAT
6
Owner
Management
Revision
Adverse Opinion
Not complying to bookkeeping act
VAT not filed on time
Adverse Opinion
Not complying to bookkeeping act
VAT not filed on time
Salery tax not paid
Adverse Opinion
Holding company
OwnerOwner
Real owner
Danish Business Authority
Et nyt dataparadigme
Erhvervsstyrelsen 7
(Legal) network of a lawyer
with roles in relation to 12.400 companies
Transformation
8Danish Business Authority
What’s the deal with Graph and ML?
 ML is based on data properties, but isn’t suited to handle
relations between objects in data
 Graph provides context to ML and even supports algorithms
based on data structure
9Danish Business Authority
Currently 126 mio. nodes
160 mio. relations
ML insights persisted to graph
10Danish Business Authority
Blue: Company
Yellow: Person
Purple: Annual report
Red: ML insights
Machine learning
controls all identity
papers for foreign
business actors
ML controls that
fictional assets are
not inserted
‘Weaponize’
unstructured data
concerning
negligence
Control new
businesses for
concerns of fraud
Identity
Assets
Audits
1.st line
Handling complexity
- 4 intelligent controls in 2019
Erhvervsstyrelsen 11
Connected
data from data
Erhvervsstyrelsen 12
Erhvervsstyrelsen 13
Registry data
Business registry
VAT
Annual reports
data
Erhvervsstyrelsen 14
Registry data
Network
data
Erhvervsstyrelsen 15
Registry data + metadata
Data from data
Delta values
Discrepancies
Client profile, IP, timestamp
data
metadata
Erhvervsstyrelsen 16
Registry data + metadata
Enriched network
data
metadata
Erhvervsstyrelsen 17
Registry data + metadata + observations
Shares client
Group
Fictionous
Anormalities
data
metadata
Machine learning
Erhvervsstyrelsen 18
Registry data + metadata + observations
Shares client
Group
Fictionous
Anormalities
data
metadata
Machine learning
Erhvervsstyrelsen 19
data
metadata
Machine learning
Registry data + metadata + observations
Erhvervsstyrelsen 20
Automatic control of new data
Exploits what we already know
Uses machine insights
Machine learning
Registry data + metadata + observations
Erhvervsstyrelsen 21
Data from data growth
Data Metadata ML Automate
01 02 03 04
Information about
persons,
companies, annual
reports, VAT etc.
Data from data. Observations,
machine driven
insights.
Data driven
business.
Registries Metadata ML Business
Intelligent control
Erhvervsstyrelsen 22
ERST ML data platform
Erhvervsstyrelsen 23
Machine learning models
use and enrich our
Knowledge graph
triggered by events in near
real time
Knowledge graph maintains
360° network analysis of
customers and business life
cycles
ML data platform
Cloud infrastructure
Event driven architecture
ML data governance
Data event store
Automated intelligent controls applied to
business systems in support of decision making.
What is complicated?
 ML data governance
 Machine learning in production
 Reacting in near real-time
 Business transformation
 Explainability
 Automation
24Danish Business Authority
Transparency and fairness in AI
 Data ethics
Erhvervsstyrelsen 25
Traceability in data
26
Business
Who did what?
Technology
Data lineage, metadata management
Evaluation
Can we do better?
Danish Business Authority
The knowledge graph and semantic AI
Erhvervsstyrelsen 27
Graph as a knowledge catalyst
28Danish Business Authority
Data sources
Meta model
Agent
ML enrichment
Knowledge graph
Automation
Semantic AI
EVENT DATA
The semantic journey
29Danish Business Authority
Data sources
Meta model
Agent
ML enrichment
Knowledge graph
Automation
Semantic AI
Knowledge AI
30Danish Business Authority
AI abstraction
Semantic layer
The principles
 Graph adoption to contextualize business lifecycles
 Meta data strategy to produce data from data
 ML enriched automation so we may adopt machine generated insight
 Monitor and trace usage so we can explain
 Evaluate and improve continuously
Erhvervsstyrelsen 31
Questions?
32Danish Business Authority
Marius Hartmann
marhar@erst.dk
+45 35 29 19 46

More Related Content

What's hot

The Knowledge Graph Explosion
The Knowledge Graph ExplosionThe Knowledge Graph Explosion
The Knowledge Graph ExplosionNeo4j
 
Graph Databases for Master Data Management
Graph Databases for Master Data ManagementGraph Databases for Master Data Management
Graph Databases for Master Data ManagementNeo4j
 
EY: Why graph technology makes sense for fraud detection and customer 360 pro...
EY: Why graph technology makes sense for fraud detection and customer 360 pro...EY: Why graph technology makes sense for fraud detection and customer 360 pro...
EY: Why graph technology makes sense for fraud detection and customer 360 pro...Neo4j
 
How Graph Data Science can turbocharge your Knowledge Graph
How Graph Data Science can turbocharge your Knowledge GraphHow Graph Data Science can turbocharge your Knowledge Graph
How Graph Data Science can turbocharge your Knowledge GraphNeo4j
 
Neo4j Generative AI workshop at GraphSummit London 14 Nov 2023.pdf
Neo4j Generative AI workshop at GraphSummit London 14 Nov 2023.pdfNeo4j Generative AI workshop at GraphSummit London 14 Nov 2023.pdf
Neo4j Generative AI workshop at GraphSummit London 14 Nov 2023.pdfNeo4j
 
The Art of the Possible with Graph - Sudhir Hasbe - GraphSummit London 14 Nov...
The Art of the Possible with Graph - Sudhir Hasbe - GraphSummit London 14 Nov...The Art of the Possible with Graph - Sudhir Hasbe - GraphSummit London 14 Nov...
The Art of the Possible with Graph - Sudhir Hasbe - GraphSummit London 14 Nov...Neo4j
 
Graph Databases – Benefits and Risks
Graph Databases – Benefits and RisksGraph Databases – Benefits and Risks
Graph Databases – Benefits and RisksDATAVERSITY
 
EY + Neo4j: Why graph technology makes sense for fraud detection and customer...
EY + Neo4j: Why graph technology makes sense for fraud detection and customer...EY + Neo4j: Why graph technology makes sense for fraud detection and customer...
EY + Neo4j: Why graph technology makes sense for fraud detection and customer...Neo4j
 
GPT and Graph Data Science to power your Knowledge Graph
GPT and Graph Data Science to power your Knowledge GraphGPT and Graph Data Science to power your Knowledge Graph
GPT and Graph Data Science to power your Knowledge GraphNeo4j
 
Neo4j GraphSummit London - The Path To Success With Graph Database and Data S...
Neo4j GraphSummit London - The Path To Success With Graph Database and Data S...Neo4j GraphSummit London - The Path To Success With Graph Database and Data S...
Neo4j GraphSummit London - The Path To Success With Graph Database and Data S...Neo4j
 
Knowledge Graphs are Worthless, Knowledge Graph Use Cases are Priceless
Knowledge Graphs are Worthless, Knowledge Graph Use Cases are PricelessKnowledge Graphs are Worthless, Knowledge Graph Use Cases are Priceless
Knowledge Graphs are Worthless, Knowledge Graph Use Cases are PricelessEnterprise Knowledge
 
Amsterdam - The Neo4j Graph Data Platform Today & Tomorrow
Amsterdam - The Neo4j Graph Data Platform Today & TomorrowAmsterdam - The Neo4j Graph Data Platform Today & Tomorrow
Amsterdam - The Neo4j Graph Data Platform Today & TomorrowNeo4j
 
Graphs in Automotive and Manufacturing - Unlock New Value from Your Data
Graphs in Automotive and Manufacturing - Unlock New Value from Your DataGraphs in Automotive and Manufacturing - Unlock New Value from Your Data
Graphs in Automotive and Manufacturing - Unlock New Value from Your DataNeo4j
 
Knowledge Graphs & Graph Data Science, More Context, Better Predictions - Neo...
Knowledge Graphs & Graph Data Science, More Context, Better Predictions - Neo...Knowledge Graphs & Graph Data Science, More Context, Better Predictions - Neo...
Knowledge Graphs & Graph Data Science, More Context, Better Predictions - Neo...Neo4j
 
How Expedia’s Entity Graph Powers Global Travel
How Expedia’s Entity Graph Powers Global TravelHow Expedia’s Entity Graph Powers Global Travel
How Expedia’s Entity Graph Powers Global TravelNeo4j
 
How to Build a Fraud Detection Solution with Neo4j
How to Build a Fraud Detection Solution with Neo4jHow to Build a Fraud Detection Solution with Neo4j
How to Build a Fraud Detection Solution with Neo4jNeo4j
 
Emerging Trends in Data Architecture – What’s the Next Big Thing?
Emerging Trends in Data Architecture – What’s the Next Big Thing?Emerging Trends in Data Architecture – What’s the Next Big Thing?
Emerging Trends in Data Architecture – What’s the Next Big Thing?DATAVERSITY
 
Data Lineage: Using Knowledge Graphs for Deeper Insights into Your Data Pipel...
Data Lineage: Using Knowledge Graphs for Deeper Insights into Your Data Pipel...Data Lineage: Using Knowledge Graphs for Deeper Insights into Your Data Pipel...
Data Lineage: Using Knowledge Graphs for Deeper Insights into Your Data Pipel...Neo4j
 
The art of the possible with graph technology_Neo4j GraphSummit Dublin 2023.pptx
The art of the possible with graph technology_Neo4j GraphSummit Dublin 2023.pptxThe art of the possible with graph technology_Neo4j GraphSummit Dublin 2023.pptx
The art of the possible with graph technology_Neo4j GraphSummit Dublin 2023.pptxNeo4j
 
Banking Circle: Money Laundering Beware: A Modern Approach to AML with Machin...
Banking Circle: Money Laundering Beware: A Modern Approach to AML with Machin...Banking Circle: Money Laundering Beware: A Modern Approach to AML with Machin...
Banking Circle: Money Laundering Beware: A Modern Approach to AML with Machin...Neo4j
 

What's hot (20)

The Knowledge Graph Explosion
The Knowledge Graph ExplosionThe Knowledge Graph Explosion
The Knowledge Graph Explosion
 
Graph Databases for Master Data Management
Graph Databases for Master Data ManagementGraph Databases for Master Data Management
Graph Databases for Master Data Management
 
EY: Why graph technology makes sense for fraud detection and customer 360 pro...
EY: Why graph technology makes sense for fraud detection and customer 360 pro...EY: Why graph technology makes sense for fraud detection and customer 360 pro...
EY: Why graph technology makes sense for fraud detection and customer 360 pro...
 
How Graph Data Science can turbocharge your Knowledge Graph
How Graph Data Science can turbocharge your Knowledge GraphHow Graph Data Science can turbocharge your Knowledge Graph
How Graph Data Science can turbocharge your Knowledge Graph
 
Neo4j Generative AI workshop at GraphSummit London 14 Nov 2023.pdf
Neo4j Generative AI workshop at GraphSummit London 14 Nov 2023.pdfNeo4j Generative AI workshop at GraphSummit London 14 Nov 2023.pdf
Neo4j Generative AI workshop at GraphSummit London 14 Nov 2023.pdf
 
The Art of the Possible with Graph - Sudhir Hasbe - GraphSummit London 14 Nov...
The Art of the Possible with Graph - Sudhir Hasbe - GraphSummit London 14 Nov...The Art of the Possible with Graph - Sudhir Hasbe - GraphSummit London 14 Nov...
The Art of the Possible with Graph - Sudhir Hasbe - GraphSummit London 14 Nov...
 
Graph Databases – Benefits and Risks
Graph Databases – Benefits and RisksGraph Databases – Benefits and Risks
Graph Databases – Benefits and Risks
 
EY + Neo4j: Why graph technology makes sense for fraud detection and customer...
EY + Neo4j: Why graph technology makes sense for fraud detection and customer...EY + Neo4j: Why graph technology makes sense for fraud detection and customer...
EY + Neo4j: Why graph technology makes sense for fraud detection and customer...
 
GPT and Graph Data Science to power your Knowledge Graph
GPT and Graph Data Science to power your Knowledge GraphGPT and Graph Data Science to power your Knowledge Graph
GPT and Graph Data Science to power your Knowledge Graph
 
Neo4j GraphSummit London - The Path To Success With Graph Database and Data S...
Neo4j GraphSummit London - The Path To Success With Graph Database and Data S...Neo4j GraphSummit London - The Path To Success With Graph Database and Data S...
Neo4j GraphSummit London - The Path To Success With Graph Database and Data S...
 
Knowledge Graphs are Worthless, Knowledge Graph Use Cases are Priceless
Knowledge Graphs are Worthless, Knowledge Graph Use Cases are PricelessKnowledge Graphs are Worthless, Knowledge Graph Use Cases are Priceless
Knowledge Graphs are Worthless, Knowledge Graph Use Cases are Priceless
 
Amsterdam - The Neo4j Graph Data Platform Today & Tomorrow
Amsterdam - The Neo4j Graph Data Platform Today & TomorrowAmsterdam - The Neo4j Graph Data Platform Today & Tomorrow
Amsterdam - The Neo4j Graph Data Platform Today & Tomorrow
 
Graphs in Automotive and Manufacturing - Unlock New Value from Your Data
Graphs in Automotive and Manufacturing - Unlock New Value from Your DataGraphs in Automotive and Manufacturing - Unlock New Value from Your Data
Graphs in Automotive and Manufacturing - Unlock New Value from Your Data
 
Knowledge Graphs & Graph Data Science, More Context, Better Predictions - Neo...
Knowledge Graphs & Graph Data Science, More Context, Better Predictions - Neo...Knowledge Graphs & Graph Data Science, More Context, Better Predictions - Neo...
Knowledge Graphs & Graph Data Science, More Context, Better Predictions - Neo...
 
How Expedia’s Entity Graph Powers Global Travel
How Expedia’s Entity Graph Powers Global TravelHow Expedia’s Entity Graph Powers Global Travel
How Expedia’s Entity Graph Powers Global Travel
 
How to Build a Fraud Detection Solution with Neo4j
How to Build a Fraud Detection Solution with Neo4jHow to Build a Fraud Detection Solution with Neo4j
How to Build a Fraud Detection Solution with Neo4j
 
Emerging Trends in Data Architecture – What’s the Next Big Thing?
Emerging Trends in Data Architecture – What’s the Next Big Thing?Emerging Trends in Data Architecture – What’s the Next Big Thing?
Emerging Trends in Data Architecture – What’s the Next Big Thing?
 
Data Lineage: Using Knowledge Graphs for Deeper Insights into Your Data Pipel...
Data Lineage: Using Knowledge Graphs for Deeper Insights into Your Data Pipel...Data Lineage: Using Knowledge Graphs for Deeper Insights into Your Data Pipel...
Data Lineage: Using Knowledge Graphs for Deeper Insights into Your Data Pipel...
 
The art of the possible with graph technology_Neo4j GraphSummit Dublin 2023.pptx
The art of the possible with graph technology_Neo4j GraphSummit Dublin 2023.pptxThe art of the possible with graph technology_Neo4j GraphSummit Dublin 2023.pptx
The art of the possible with graph technology_Neo4j GraphSummit Dublin 2023.pptx
 
Banking Circle: Money Laundering Beware: A Modern Approach to AML with Machin...
Banking Circle: Money Laundering Beware: A Modern Approach to AML with Machin...Banking Circle: Money Laundering Beware: A Modern Approach to AML with Machin...
Banking Circle: Money Laundering Beware: A Modern Approach to AML with Machin...
 

Similar to GraphTalk Copenhagen - Fraud Detection with Graphs

Digital Transformation | The changing relationship between accountants and th...
Digital Transformation | The changing relationship between accountants and th...Digital Transformation | The changing relationship between accountants and th...
Digital Transformation | The changing relationship between accountants and th...Rebecca Hallworth
 
FinTech Belgium Summit 2018 - Vadis Technologies - Jean-François Heering
FinTech Belgium Summit 2018 - Vadis Technologies - Jean-François HeeringFinTech Belgium Summit 2018 - Vadis Technologies - Jean-François Heering
FinTech Belgium Summit 2018 - Vadis Technologies - Jean-François HeeringFinTech Belgium
 
FinTech Belgium - MeetUp on The Ideal RegTech for Banks and FinTechs - Jean-F...
FinTech Belgium - MeetUp on The Ideal RegTech for Banks and FinTechs - Jean-F...FinTech Belgium - MeetUp on The Ideal RegTech for Banks and FinTechs - Jean-F...
FinTech Belgium - MeetUp on The Ideal RegTech for Banks and FinTechs - Jean-F...FinTech Belgium
 
AXA x DSSG Meetup Sharing (Feb 2016)
AXA x DSSG Meetup Sharing (Feb 2016)AXA x DSSG Meetup Sharing (Feb 2016)
AXA x DSSG Meetup Sharing (Feb 2016)Eugene Yan Ziyou
 
DN18 | Deploying a Risk Monitoring Tool on Third Parties | Jean-François Heer...
DN18 | Deploying a Risk Monitoring Tool on Third Parties | Jean-François Heer...DN18 | Deploying a Risk Monitoring Tool on Third Parties | Jean-François Heer...
DN18 | Deploying a Risk Monitoring Tool on Third Parties | Jean-François Heer...Dataconomy Media
 
Big Data- The Next Big Thing In Accounting
Big Data- The Next Big Thing In AccountingBig Data- The Next Big Thing In Accounting
Big Data- The Next Big Thing In AccountingPooja Dhingra
 
Preventing Tax Evasion & Benefits Fraud Through Predictive Analytics
Preventing Tax Evasion & Benefits Fraud Through Predictive AnalyticsPreventing Tax Evasion & Benefits Fraud Through Predictive Analytics
Preventing Tax Evasion & Benefits Fraud Through Predictive AnalyticsCapgemini
 
How can Technology support Global Compliance and Risk Management?
How can Technology support Global Compliance and Risk Management?How can Technology support Global Compliance and Risk Management?
How can Technology support Global Compliance and Risk Management?Pierre Arman
 
Startup InsurTech Award - Digital Fineprint
Startup InsurTech Award - Digital FineprintStartup InsurTech Award - Digital Fineprint
Startup InsurTech Award - Digital FineprintThe Digital Insurer
 
Digital Fineprint Introduction deck
Digital Fineprint Introduction deckDigital Fineprint Introduction deck
Digital Fineprint Introduction deckAnna Kurmanbaeva
 
These Fascinating Examples Show Why Streaming Data And Real-Time Analytics Ma...
These Fascinating Examples Show Why Streaming Data And Real-Time Analytics Ma...These Fascinating Examples Show Why Streaming Data And Real-Time Analytics Ma...
These Fascinating Examples Show Why Streaming Data And Real-Time Analytics Ma...Bernard Marr
 
Big data, data analytics and effective tax administration in nigeria
Big data, data analytics and effective tax administration in nigeriaBig data, data analytics and effective tax administration in nigeria
Big data, data analytics and effective tax administration in nigeriaJoeBrands1
 
Protect Your Revenue Streams: Big Data & Analytics in Tax
Protect Your Revenue Streams: Big Data & Analytics in TaxProtect Your Revenue Streams: Big Data & Analytics in Tax
Protect Your Revenue Streams: Big Data & Analytics in TaxCapgemini
 
Capgemini’s Data WARP: Accelerate your Journey to Insights
Capgemini’s Data WARP: Accelerate your Journey to InsightsCapgemini’s Data WARP: Accelerate your Journey to Insights
Capgemini’s Data WARP: Accelerate your Journey to InsightsCapgemini
 
Big Data: Real-life examples of Business Value Generation with Cloudera
Big Data: Real-life examples of Business Value Generation with ClouderaBig Data: Real-life examples of Business Value Generation with Cloudera
Big Data: Real-life examples of Business Value Generation with ClouderaCapgemini
 
Big Data and MDM altogether: the winning association
Big Data and MDM altogether: the winning associationBig Data and MDM altogether: the winning association
Big Data and MDM altogether: the winning associationJean-Michel Franco
 
Gdpr compliance. Presentation for Consulegis Lawyers network
Gdpr compliance.  Presentation  for Consulegis Lawyers networkGdpr compliance.  Presentation  for Consulegis Lawyers network
Gdpr compliance. Presentation for Consulegis Lawyers networkBart Van Den Brande
 

Similar to GraphTalk Copenhagen - Fraud Detection with Graphs (20)

Digital Transformation | The changing relationship between accountants and th...
Digital Transformation | The changing relationship between accountants and th...Digital Transformation | The changing relationship between accountants and th...
Digital Transformation | The changing relationship between accountants and th...
 
FinTech Belgium Summit 2018 - Vadis Technologies - Jean-François Heering
FinTech Belgium Summit 2018 - Vadis Technologies - Jean-François HeeringFinTech Belgium Summit 2018 - Vadis Technologies - Jean-François Heering
FinTech Belgium Summit 2018 - Vadis Technologies - Jean-François Heering
 
FinTech Belgium - MeetUp on The Ideal RegTech for Banks and FinTechs - Jean-F...
FinTech Belgium - MeetUp on The Ideal RegTech for Banks and FinTechs - Jean-F...FinTech Belgium - MeetUp on The Ideal RegTech for Banks and FinTechs - Jean-F...
FinTech Belgium - MeetUp on The Ideal RegTech for Banks and FinTechs - Jean-F...
 
Peta Pilot
Peta PilotPeta Pilot
Peta Pilot
 
AXA x DSSG Meetup Sharing (Feb 2016)
AXA x DSSG Meetup Sharing (Feb 2016)AXA x DSSG Meetup Sharing (Feb 2016)
AXA x DSSG Meetup Sharing (Feb 2016)
 
DN18 | Deploying a Risk Monitoring Tool on Third Parties | Jean-François Heer...
DN18 | Deploying a Risk Monitoring Tool on Third Parties | Jean-François Heer...DN18 | Deploying a Risk Monitoring Tool on Third Parties | Jean-François Heer...
DN18 | Deploying a Risk Monitoring Tool on Third Parties | Jean-François Heer...
 
Big Data- The Next Big Thing In Accounting
Big Data- The Next Big Thing In AccountingBig Data- The Next Big Thing In Accounting
Big Data- The Next Big Thing In Accounting
 
Preventing Tax Evasion & Benefits Fraud Through Predictive Analytics
Preventing Tax Evasion & Benefits Fraud Through Predictive AnalyticsPreventing Tax Evasion & Benefits Fraud Through Predictive Analytics
Preventing Tax Evasion & Benefits Fraud Through Predictive Analytics
 
How can Technology support Global Compliance and Risk Management?
How can Technology support Global Compliance and Risk Management?How can Technology support Global Compliance and Risk Management?
How can Technology support Global Compliance and Risk Management?
 
ADMINEX
ADMINEXADMINEX
ADMINEX
 
Startup InsurTech Award - Digital Fineprint
Startup InsurTech Award - Digital FineprintStartup InsurTech Award - Digital Fineprint
Startup InsurTech Award - Digital Fineprint
 
Digital Fineprint Introduction deck
Digital Fineprint Introduction deckDigital Fineprint Introduction deck
Digital Fineprint Introduction deck
 
These Fascinating Examples Show Why Streaming Data And Real-Time Analytics Ma...
These Fascinating Examples Show Why Streaming Data And Real-Time Analytics Ma...These Fascinating Examples Show Why Streaming Data And Real-Time Analytics Ma...
These Fascinating Examples Show Why Streaming Data And Real-Time Analytics Ma...
 
Big data, data analytics and effective tax administration in nigeria
Big data, data analytics and effective tax administration in nigeriaBig data, data analytics and effective tax administration in nigeria
Big data, data analytics and effective tax administration in nigeria
 
Protect Your Revenue Streams: Big Data & Analytics in Tax
Protect Your Revenue Streams: Big Data & Analytics in TaxProtect Your Revenue Streams: Big Data & Analytics in Tax
Protect Your Revenue Streams: Big Data & Analytics in Tax
 
Capgemini’s Data WARP: Accelerate your Journey to Insights
Capgemini’s Data WARP: Accelerate your Journey to InsightsCapgemini’s Data WARP: Accelerate your Journey to Insights
Capgemini’s Data WARP: Accelerate your Journey to Insights
 
Big Data: Real-life examples of Business Value Generation with Cloudera
Big Data: Real-life examples of Business Value Generation with ClouderaBig Data: Real-life examples of Business Value Generation with Cloudera
Big Data: Real-life examples of Business Value Generation with Cloudera
 
Big Data and MDM altogether: the winning association
Big Data and MDM altogether: the winning associationBig Data and MDM altogether: the winning association
Big Data and MDM altogether: the winning association
 
Gdpr compliance. Presentation for Consulegis Lawyers network
Gdpr compliance.  Presentation  for Consulegis Lawyers networkGdpr compliance.  Presentation  for Consulegis Lawyers network
Gdpr compliance. Presentation for Consulegis Lawyers network
 
The Face of the New Enterprise
The Face of the New EnterpriseThe Face of the New Enterprise
The Face of the New Enterprise
 

More from Neo4j

Workshop - Best of Both Worlds_ Combine KG and Vector search for enhanced R...
Workshop - Best of Both Worlds_ Combine  KG and Vector search for  enhanced R...Workshop - Best of Both Worlds_ Combine  KG and Vector search for  enhanced R...
Workshop - Best of Both Worlds_ Combine KG and Vector search for enhanced R...Neo4j
 
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...Neo4j
 
QIAGEN: Biomedical Knowledge Graphs for Data Scientists and Bioinformaticians
QIAGEN: Biomedical Knowledge Graphs for Data Scientists and BioinformaticiansQIAGEN: Biomedical Knowledge Graphs for Data Scientists and Bioinformaticians
QIAGEN: Biomedical Knowledge Graphs for Data Scientists and BioinformaticiansNeo4j
 
EY_Graph Database Powered Sustainability
EY_Graph Database Powered SustainabilityEY_Graph Database Powered Sustainability
EY_Graph Database Powered SustainabilityNeo4j
 
SIEMENS: RAPUNZEL – A Tale About Knowledge Graph
SIEMENS: RAPUNZEL – A Tale About Knowledge GraphSIEMENS: RAPUNZEL – A Tale About Knowledge Graph
SIEMENS: RAPUNZEL – A Tale About Knowledge GraphNeo4j
 
Build your next Gen AI Breakthrough - April 2024
Build your next Gen AI Breakthrough - April 2024Build your next Gen AI Breakthrough - April 2024
Build your next Gen AI Breakthrough - April 2024Neo4j
 
Connecting the Dots for Information Discovery.pdf
Connecting the Dots for Information Discovery.pdfConnecting the Dots for Information Discovery.pdf
Connecting the Dots for Information Discovery.pdfNeo4j
 
ISDEFE - GraphSummit Madrid - ARETA: Aviation Real-Time Emissions Token Accre...
ISDEFE - GraphSummit Madrid - ARETA: Aviation Real-Time Emissions Token Accre...ISDEFE - GraphSummit Madrid - ARETA: Aviation Real-Time Emissions Token Accre...
ISDEFE - GraphSummit Madrid - ARETA: Aviation Real-Time Emissions Token Accre...Neo4j
 
BBVA - GraphSummit Madrid - Caso de éxito en BBVA: Optimizando con grafos
BBVA - GraphSummit Madrid - Caso de éxito en BBVA: Optimizando con grafosBBVA - GraphSummit Madrid - Caso de éxito en BBVA: Optimizando con grafos
BBVA - GraphSummit Madrid - Caso de éxito en BBVA: Optimizando con grafosNeo4j
 
Graph Everywhere - Josep Taruella - Por qué Graph Data Science en tus modelos...
Graph Everywhere - Josep Taruella - Por qué Graph Data Science en tus modelos...Graph Everywhere - Josep Taruella - Por qué Graph Data Science en tus modelos...
Graph Everywhere - Josep Taruella - Por qué Graph Data Science en tus modelos...Neo4j
 
GraphSummit Madrid - Product Vision and Roadmap - Luis Salvador Neo4j
GraphSummit Madrid - Product Vision and Roadmap - Luis Salvador Neo4jGraphSummit Madrid - Product Vision and Roadmap - Luis Salvador Neo4j
GraphSummit Madrid - Product Vision and Roadmap - Luis Salvador Neo4jNeo4j
 
Neo4j_Exploring the Impact of Graph Technology on Financial Services.pdf
Neo4j_Exploring the Impact of Graph Technology on Financial Services.pdfNeo4j_Exploring the Impact of Graph Technology on Financial Services.pdf
Neo4j_Exploring the Impact of Graph Technology on Financial Services.pdfNeo4j
 
Rabobank_Exploring the Impact of Graph Technology on Financial Services.pdf
Rabobank_Exploring the Impact of Graph Technology on Financial Services.pdfRabobank_Exploring the Impact of Graph Technology on Financial Services.pdf
Rabobank_Exploring the Impact of Graph Technology on Financial Services.pdfNeo4j
 
Webinar - IA generativa e grafi Neo4j: RAG time!
Webinar - IA generativa e grafi Neo4j: RAG time!Webinar - IA generativa e grafi Neo4j: RAG time!
Webinar - IA generativa e grafi Neo4j: RAG time!Neo4j
 
IA Generativa y Grafos de Neo4j: RAG time
IA Generativa y Grafos de Neo4j: RAG timeIA Generativa y Grafos de Neo4j: RAG time
IA Generativa y Grafos de Neo4j: RAG timeNeo4j
 
Neo4j: Data Engineering for RAG (retrieval augmented generation)
Neo4j: Data Engineering for RAG (retrieval augmented generation)Neo4j: Data Engineering for RAG (retrieval augmented generation)
Neo4j: Data Engineering for RAG (retrieval augmented generation)Neo4j
 
Neo4j Graph Summit 2024 Workshop - EMEA - Breda_and_Munchen.pdf
Neo4j Graph Summit 2024 Workshop - EMEA - Breda_and_Munchen.pdfNeo4j Graph Summit 2024 Workshop - EMEA - Breda_and_Munchen.pdf
Neo4j Graph Summit 2024 Workshop - EMEA - Breda_and_Munchen.pdfNeo4j
 
Enabling GenAI Breakthroughs with Knowledge Graphs
Enabling GenAI Breakthroughs with Knowledge GraphsEnabling GenAI Breakthroughs with Knowledge Graphs
Enabling GenAI Breakthroughs with Knowledge GraphsNeo4j
 
Neo4j_Anurag Tandon_Product Vision and Roadmap.Benelux.pptx.pdf
Neo4j_Anurag Tandon_Product Vision and Roadmap.Benelux.pptx.pdfNeo4j_Anurag Tandon_Product Vision and Roadmap.Benelux.pptx.pdf
Neo4j_Anurag Tandon_Product Vision and Roadmap.Benelux.pptx.pdfNeo4j
 
Neo4j Jesus Barrasa The Art of the Possible with Graph
Neo4j Jesus Barrasa The Art of the Possible with GraphNeo4j Jesus Barrasa The Art of the Possible with Graph
Neo4j Jesus Barrasa The Art of the Possible with GraphNeo4j
 

More from Neo4j (20)

Workshop - Best of Both Worlds_ Combine KG and Vector search for enhanced R...
Workshop - Best of Both Worlds_ Combine  KG and Vector search for  enhanced R...Workshop - Best of Both Worlds_ Combine  KG and Vector search for  enhanced R...
Workshop - Best of Both Worlds_ Combine KG and Vector search for enhanced R...
 
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
 
QIAGEN: Biomedical Knowledge Graphs for Data Scientists and Bioinformaticians
QIAGEN: Biomedical Knowledge Graphs for Data Scientists and BioinformaticiansQIAGEN: Biomedical Knowledge Graphs for Data Scientists and Bioinformaticians
QIAGEN: Biomedical Knowledge Graphs for Data Scientists and Bioinformaticians
 
EY_Graph Database Powered Sustainability
EY_Graph Database Powered SustainabilityEY_Graph Database Powered Sustainability
EY_Graph Database Powered Sustainability
 
SIEMENS: RAPUNZEL – A Tale About Knowledge Graph
SIEMENS: RAPUNZEL – A Tale About Knowledge GraphSIEMENS: RAPUNZEL – A Tale About Knowledge Graph
SIEMENS: RAPUNZEL – A Tale About Knowledge Graph
 
Build your next Gen AI Breakthrough - April 2024
Build your next Gen AI Breakthrough - April 2024Build your next Gen AI Breakthrough - April 2024
Build your next Gen AI Breakthrough - April 2024
 
Connecting the Dots for Information Discovery.pdf
Connecting the Dots for Information Discovery.pdfConnecting the Dots for Information Discovery.pdf
Connecting the Dots for Information Discovery.pdf
 
ISDEFE - GraphSummit Madrid - ARETA: Aviation Real-Time Emissions Token Accre...
ISDEFE - GraphSummit Madrid - ARETA: Aviation Real-Time Emissions Token Accre...ISDEFE - GraphSummit Madrid - ARETA: Aviation Real-Time Emissions Token Accre...
ISDEFE - GraphSummit Madrid - ARETA: Aviation Real-Time Emissions Token Accre...
 
BBVA - GraphSummit Madrid - Caso de éxito en BBVA: Optimizando con grafos
BBVA - GraphSummit Madrid - Caso de éxito en BBVA: Optimizando con grafosBBVA - GraphSummit Madrid - Caso de éxito en BBVA: Optimizando con grafos
BBVA - GraphSummit Madrid - Caso de éxito en BBVA: Optimizando con grafos
 
Graph Everywhere - Josep Taruella - Por qué Graph Data Science en tus modelos...
Graph Everywhere - Josep Taruella - Por qué Graph Data Science en tus modelos...Graph Everywhere - Josep Taruella - Por qué Graph Data Science en tus modelos...
Graph Everywhere - Josep Taruella - Por qué Graph Data Science en tus modelos...
 
GraphSummit Madrid - Product Vision and Roadmap - Luis Salvador Neo4j
GraphSummit Madrid - Product Vision and Roadmap - Luis Salvador Neo4jGraphSummit Madrid - Product Vision and Roadmap - Luis Salvador Neo4j
GraphSummit Madrid - Product Vision and Roadmap - Luis Salvador Neo4j
 
Neo4j_Exploring the Impact of Graph Technology on Financial Services.pdf
Neo4j_Exploring the Impact of Graph Technology on Financial Services.pdfNeo4j_Exploring the Impact of Graph Technology on Financial Services.pdf
Neo4j_Exploring the Impact of Graph Technology on Financial Services.pdf
 
Rabobank_Exploring the Impact of Graph Technology on Financial Services.pdf
Rabobank_Exploring the Impact of Graph Technology on Financial Services.pdfRabobank_Exploring the Impact of Graph Technology on Financial Services.pdf
Rabobank_Exploring the Impact of Graph Technology on Financial Services.pdf
 
Webinar - IA generativa e grafi Neo4j: RAG time!
Webinar - IA generativa e grafi Neo4j: RAG time!Webinar - IA generativa e grafi Neo4j: RAG time!
Webinar - IA generativa e grafi Neo4j: RAG time!
 
IA Generativa y Grafos de Neo4j: RAG time
IA Generativa y Grafos de Neo4j: RAG timeIA Generativa y Grafos de Neo4j: RAG time
IA Generativa y Grafos de Neo4j: RAG time
 
Neo4j: Data Engineering for RAG (retrieval augmented generation)
Neo4j: Data Engineering for RAG (retrieval augmented generation)Neo4j: Data Engineering for RAG (retrieval augmented generation)
Neo4j: Data Engineering for RAG (retrieval augmented generation)
 
Neo4j Graph Summit 2024 Workshop - EMEA - Breda_and_Munchen.pdf
Neo4j Graph Summit 2024 Workshop - EMEA - Breda_and_Munchen.pdfNeo4j Graph Summit 2024 Workshop - EMEA - Breda_and_Munchen.pdf
Neo4j Graph Summit 2024 Workshop - EMEA - Breda_and_Munchen.pdf
 
Enabling GenAI Breakthroughs with Knowledge Graphs
Enabling GenAI Breakthroughs with Knowledge GraphsEnabling GenAI Breakthroughs with Knowledge Graphs
Enabling GenAI Breakthroughs with Knowledge Graphs
 
Neo4j_Anurag Tandon_Product Vision and Roadmap.Benelux.pptx.pdf
Neo4j_Anurag Tandon_Product Vision and Roadmap.Benelux.pptx.pdfNeo4j_Anurag Tandon_Product Vision and Roadmap.Benelux.pptx.pdf
Neo4j_Anurag Tandon_Product Vision and Roadmap.Benelux.pptx.pdf
 
Neo4j Jesus Barrasa The Art of the Possible with Graph
Neo4j Jesus Barrasa The Art of the Possible with GraphNeo4j Jesus Barrasa The Art of the Possible with Graph
Neo4j Jesus Barrasa The Art of the Possible with Graph
 

Recently uploaded

5 Signs You Need a Fashion PLM Software.pdf
5 Signs You Need a Fashion PLM Software.pdf5 Signs You Need a Fashion PLM Software.pdf
5 Signs You Need a Fashion PLM Software.pdfWave PLM
 
How To Troubleshoot Collaboration Apps for the Modern Connected Worker
How To Troubleshoot Collaboration Apps for the Modern Connected WorkerHow To Troubleshoot Collaboration Apps for the Modern Connected Worker
How To Troubleshoot Collaboration Apps for the Modern Connected WorkerThousandEyes
 
Professional Resume Template for Software Developers
Professional Resume Template for Software DevelopersProfessional Resume Template for Software Developers
Professional Resume Template for Software DevelopersVinodh Ram
 
Unlocking the Future of AI Agents with Large Language Models
Unlocking the Future of AI Agents with Large Language ModelsUnlocking the Future of AI Agents with Large Language Models
Unlocking the Future of AI Agents with Large Language Modelsaagamshah0812
 
Tech Tuesday-Harness the Power of Effective Resource Planning with OnePlan’s ...
Tech Tuesday-Harness the Power of Effective Resource Planning with OnePlan’s ...Tech Tuesday-Harness the Power of Effective Resource Planning with OnePlan’s ...
Tech Tuesday-Harness the Power of Effective Resource Planning with OnePlan’s ...OnePlan Solutions
 
Clustering techniques data mining book ....
Clustering techniques data mining book ....Clustering techniques data mining book ....
Clustering techniques data mining book ....ShaimaaMohamedGalal
 
why an Opensea Clone Script might be your perfect match.pdf
why an Opensea Clone Script might be your perfect match.pdfwhy an Opensea Clone Script might be your perfect match.pdf
why an Opensea Clone Script might be your perfect match.pdfjoe51371421
 
W01_panagenda_Navigating-the-Future-with-The-Hitchhikers-Guide-to-Notes-and-D...
W01_panagenda_Navigating-the-Future-with-The-Hitchhikers-Guide-to-Notes-and-D...W01_panagenda_Navigating-the-Future-with-The-Hitchhikers-Guide-to-Notes-and-D...
W01_panagenda_Navigating-the-Future-with-The-Hitchhikers-Guide-to-Notes-and-D...panagenda
 
Test Automation Strategy for Frontend and Backend
Test Automation Strategy for Frontend and BackendTest Automation Strategy for Frontend and Backend
Test Automation Strategy for Frontend and BackendArshad QA
 
Short Story: Unveiling the Reasoning Abilities of Large Language Models by Ke...
Short Story: Unveiling the Reasoning Abilities of Large Language Models by Ke...Short Story: Unveiling the Reasoning Abilities of Large Language Models by Ke...
Short Story: Unveiling the Reasoning Abilities of Large Language Models by Ke...kellynguyen01
 
SyndBuddy AI 2k Review 2024: Revolutionizing Content Syndication with AI
SyndBuddy AI 2k Review 2024: Revolutionizing Content Syndication with AISyndBuddy AI 2k Review 2024: Revolutionizing Content Syndication with AI
SyndBuddy AI 2k Review 2024: Revolutionizing Content Syndication with AIABDERRAOUF MEHENNI
 
Reassessing the Bedrock of Clinical Function Models: An Examination of Large ...
Reassessing the Bedrock of Clinical Function Models: An Examination of Large ...Reassessing the Bedrock of Clinical Function Models: An Examination of Large ...
Reassessing the Bedrock of Clinical Function Models: An Examination of Large ...harshavardhanraghave
 
Software Quality Assurance Interview Questions
Software Quality Assurance Interview QuestionsSoftware Quality Assurance Interview Questions
Software Quality Assurance Interview QuestionsArshad QA
 
Optimizing AI for immediate response in Smart CCTV
Optimizing AI for immediate response in Smart CCTVOptimizing AI for immediate response in Smart CCTV
Optimizing AI for immediate response in Smart CCTVshikhaohhpro
 
The Real-World Challenges of Medical Device Cybersecurity- Mitigating Vulnera...
The Real-World Challenges of Medical Device Cybersecurity- Mitigating Vulnera...The Real-World Challenges of Medical Device Cybersecurity- Mitigating Vulnera...
The Real-World Challenges of Medical Device Cybersecurity- Mitigating Vulnera...ICS
 
Learn the Fundamentals of XCUITest Framework_ A Beginner's Guide.pdf
Learn the Fundamentals of XCUITest Framework_ A Beginner's Guide.pdfLearn the Fundamentals of XCUITest Framework_ A Beginner's Guide.pdf
Learn the Fundamentals of XCUITest Framework_ A Beginner's Guide.pdfkalichargn70th171
 
Try MyIntelliAccount Cloud Accounting Software As A Service Solution Risk Fre...
Try MyIntelliAccount Cloud Accounting Software As A Service Solution Risk Fre...Try MyIntelliAccount Cloud Accounting Software As A Service Solution Risk Fre...
Try MyIntelliAccount Cloud Accounting Software As A Service Solution Risk Fre...MyIntelliSource, Inc.
 
Hand gesture recognition PROJECT PPT.pptx
Hand gesture recognition PROJECT PPT.pptxHand gesture recognition PROJECT PPT.pptx
Hand gesture recognition PROJECT PPT.pptxbodapatigopi8531
 

Recently uploaded (20)

5 Signs You Need a Fashion PLM Software.pdf
5 Signs You Need a Fashion PLM Software.pdf5 Signs You Need a Fashion PLM Software.pdf
5 Signs You Need a Fashion PLM Software.pdf
 
How To Troubleshoot Collaboration Apps for the Modern Connected Worker
How To Troubleshoot Collaboration Apps for the Modern Connected WorkerHow To Troubleshoot Collaboration Apps for the Modern Connected Worker
How To Troubleshoot Collaboration Apps for the Modern Connected Worker
 
Professional Resume Template for Software Developers
Professional Resume Template for Software DevelopersProfessional Resume Template for Software Developers
Professional Resume Template for Software Developers
 
Unlocking the Future of AI Agents with Large Language Models
Unlocking the Future of AI Agents with Large Language ModelsUnlocking the Future of AI Agents with Large Language Models
Unlocking the Future of AI Agents with Large Language Models
 
Tech Tuesday-Harness the Power of Effective Resource Planning with OnePlan’s ...
Tech Tuesday-Harness the Power of Effective Resource Planning with OnePlan’s ...Tech Tuesday-Harness the Power of Effective Resource Planning with OnePlan’s ...
Tech Tuesday-Harness the Power of Effective Resource Planning with OnePlan’s ...
 
Clustering techniques data mining book ....
Clustering techniques data mining book ....Clustering techniques data mining book ....
Clustering techniques data mining book ....
 
why an Opensea Clone Script might be your perfect match.pdf
why an Opensea Clone Script might be your perfect match.pdfwhy an Opensea Clone Script might be your perfect match.pdf
why an Opensea Clone Script might be your perfect match.pdf
 
W01_panagenda_Navigating-the-Future-with-The-Hitchhikers-Guide-to-Notes-and-D...
W01_panagenda_Navigating-the-Future-with-The-Hitchhikers-Guide-to-Notes-and-D...W01_panagenda_Navigating-the-Future-with-The-Hitchhikers-Guide-to-Notes-and-D...
W01_panagenda_Navigating-the-Future-with-The-Hitchhikers-Guide-to-Notes-and-D...
 
Vip Call Girls Noida ➡️ Delhi ➡️ 9999965857 No Advance 24HRS Live
Vip Call Girls Noida ➡️ Delhi ➡️ 9999965857 No Advance 24HRS LiveVip Call Girls Noida ➡️ Delhi ➡️ 9999965857 No Advance 24HRS Live
Vip Call Girls Noida ➡️ Delhi ➡️ 9999965857 No Advance 24HRS Live
 
Microsoft AI Transformation Partner Playbook.pdf
Microsoft AI Transformation Partner Playbook.pdfMicrosoft AI Transformation Partner Playbook.pdf
Microsoft AI Transformation Partner Playbook.pdf
 
Test Automation Strategy for Frontend and Backend
Test Automation Strategy for Frontend and BackendTest Automation Strategy for Frontend and Backend
Test Automation Strategy for Frontend and Backend
 
Short Story: Unveiling the Reasoning Abilities of Large Language Models by Ke...
Short Story: Unveiling the Reasoning Abilities of Large Language Models by Ke...Short Story: Unveiling the Reasoning Abilities of Large Language Models by Ke...
Short Story: Unveiling the Reasoning Abilities of Large Language Models by Ke...
 
SyndBuddy AI 2k Review 2024: Revolutionizing Content Syndication with AI
SyndBuddy AI 2k Review 2024: Revolutionizing Content Syndication with AISyndBuddy AI 2k Review 2024: Revolutionizing Content Syndication with AI
SyndBuddy AI 2k Review 2024: Revolutionizing Content Syndication with AI
 
Reassessing the Bedrock of Clinical Function Models: An Examination of Large ...
Reassessing the Bedrock of Clinical Function Models: An Examination of Large ...Reassessing the Bedrock of Clinical Function Models: An Examination of Large ...
Reassessing the Bedrock of Clinical Function Models: An Examination of Large ...
 
Software Quality Assurance Interview Questions
Software Quality Assurance Interview QuestionsSoftware Quality Assurance Interview Questions
Software Quality Assurance Interview Questions
 
Optimizing AI for immediate response in Smart CCTV
Optimizing AI for immediate response in Smart CCTVOptimizing AI for immediate response in Smart CCTV
Optimizing AI for immediate response in Smart CCTV
 
The Real-World Challenges of Medical Device Cybersecurity- Mitigating Vulnera...
The Real-World Challenges of Medical Device Cybersecurity- Mitigating Vulnera...The Real-World Challenges of Medical Device Cybersecurity- Mitigating Vulnera...
The Real-World Challenges of Medical Device Cybersecurity- Mitigating Vulnera...
 
Learn the Fundamentals of XCUITest Framework_ A Beginner's Guide.pdf
Learn the Fundamentals of XCUITest Framework_ A Beginner's Guide.pdfLearn the Fundamentals of XCUITest Framework_ A Beginner's Guide.pdf
Learn the Fundamentals of XCUITest Framework_ A Beginner's Guide.pdf
 
Try MyIntelliAccount Cloud Accounting Software As A Service Solution Risk Fre...
Try MyIntelliAccount Cloud Accounting Software As A Service Solution Risk Fre...Try MyIntelliAccount Cloud Accounting Software As A Service Solution Risk Fre...
Try MyIntelliAccount Cloud Accounting Software As A Service Solution Risk Fre...
 
Hand gesture recognition PROJECT PPT.pptx
Hand gesture recognition PROJECT PPT.pptxHand gesture recognition PROJECT PPT.pptx
Hand gesture recognition PROJECT PPT.pptx
 

GraphTalk Copenhagen - Fraud Detection with Graphs

  • 1. Marius Hartmann Fraud Detection with Graphs 24. september 2019
  • 2. Main task of The Danish Business Authority 2Danish Business Authority Business service and registration of companies Business development and digital growth Business regulation and supervision planning and rural business EU and international affairs
  • 3. Virk: The joint public one-stop shop to the Danish business world 3Danish Business Authority 27.000.000 visits at Virk on an annual basis 96% of companies in Denmark know Virk 4.000.000 filings on Virk annually 92% Instant case handling
  • 4. ML Lab  9 person strong  Physics, astro-physics, economics, computer science, fine art, social science, 7 Phd’s (+1 on the way)  2/7 gender balance, kids and no-kids Erhvervsstyrelsen 4
  • 5. What can we do with ML and Graph?  Help and guide users to make fewer mistakes  Improve and scale our control and supervision  Provide recommendations and personalize our solutions  Improve our policy development with ML created insight 5Danish Business Authority
  • 6. Management Owner ? Example of control: Strengthened company control regarding VAT 6 Owner Management Revision Adverse Opinion Not complying to bookkeeping act VAT not filed on time Adverse Opinion Not complying to bookkeeping act VAT not filed on time Salery tax not paid Adverse Opinion Holding company OwnerOwner Real owner Danish Business Authority
  • 7. Et nyt dataparadigme Erhvervsstyrelsen 7 (Legal) network of a lawyer with roles in relation to 12.400 companies
  • 9. What’s the deal with Graph and ML?  ML is based on data properties, but isn’t suited to handle relations between objects in data  Graph provides context to ML and even supports algorithms based on data structure 9Danish Business Authority Currently 126 mio. nodes 160 mio. relations
  • 10. ML insights persisted to graph 10Danish Business Authority Blue: Company Yellow: Person Purple: Annual report Red: ML insights
  • 11. Machine learning controls all identity papers for foreign business actors ML controls that fictional assets are not inserted ‘Weaponize’ unstructured data concerning negligence Control new businesses for concerns of fraud Identity Assets Audits 1.st line Handling complexity - 4 intelligent controls in 2019 Erhvervsstyrelsen 11
  • 13. Erhvervsstyrelsen 13 Registry data Business registry VAT Annual reports data
  • 15. Erhvervsstyrelsen 15 Registry data + metadata Data from data Delta values Discrepancies Client profile, IP, timestamp data metadata
  • 16. Erhvervsstyrelsen 16 Registry data + metadata Enriched network data metadata
  • 17. Erhvervsstyrelsen 17 Registry data + metadata + observations Shares client Group Fictionous Anormalities data metadata Machine learning
  • 18. Erhvervsstyrelsen 18 Registry data + metadata + observations Shares client Group Fictionous Anormalities data metadata Machine learning
  • 20. Erhvervsstyrelsen 20 Automatic control of new data Exploits what we already know Uses machine insights Machine learning Registry data + metadata + observations
  • 22. Data Metadata ML Automate 01 02 03 04 Information about persons, companies, annual reports, VAT etc. Data from data. Observations, machine driven insights. Data driven business. Registries Metadata ML Business Intelligent control Erhvervsstyrelsen 22
  • 23. ERST ML data platform Erhvervsstyrelsen 23 Machine learning models use and enrich our Knowledge graph triggered by events in near real time Knowledge graph maintains 360° network analysis of customers and business life cycles ML data platform Cloud infrastructure Event driven architecture ML data governance Data event store Automated intelligent controls applied to business systems in support of decision making.
  • 24. What is complicated?  ML data governance  Machine learning in production  Reacting in near real-time  Business transformation  Explainability  Automation 24Danish Business Authority
  • 25. Transparency and fairness in AI  Data ethics Erhvervsstyrelsen 25
  • 26. Traceability in data 26 Business Who did what? Technology Data lineage, metadata management Evaluation Can we do better? Danish Business Authority
  • 27. The knowledge graph and semantic AI Erhvervsstyrelsen 27
  • 28. Graph as a knowledge catalyst 28Danish Business Authority Data sources Meta model Agent ML enrichment Knowledge graph Automation Semantic AI EVENT DATA
  • 29. The semantic journey 29Danish Business Authority Data sources Meta model Agent ML enrichment Knowledge graph Automation Semantic AI
  • 30. Knowledge AI 30Danish Business Authority AI abstraction Semantic layer
  • 31. The principles  Graph adoption to contextualize business lifecycles  Meta data strategy to produce data from data  ML enriched automation so we may adopt machine generated insight  Monitor and trace usage so we can explain  Evaluate and improve continuously Erhvervsstyrelsen 31

Editor's Notes

  1. he main tasks of The Danish Business Autority: Registration og Compagnies Business regulation and supervision Planning and rural businesse Business development and digital growth EU and international affairs We have a variety of different stakeholders; from small businesses to the large international companies. Different professional actors, the municipalities and the political system
  2. ’m very proud of these four figures. They show that Virk have truly become the public one-stop shop for businesses in Denmark. The companies know and visit Virk, and they file in their information on Virk. 92 % of cases are resolved instantly without manual processing needed.
  3. Hjælpe og vejlede: Maskinen kan f.eks. i årsrapporter læse anvendt regnskabspraksis, som er fritekst, og give brugeren en ”advis” hvis der ikke er overensstemmelse mellem på den ene side tallene i regnskabet (de mangler eller er forkerte) og på den anden side, den selvanførte regnskabspraksis. Forbedre og skalere kontrol og tilsyn: Vi vil kunne reagere allerede når en brugere forsøger at indsende noget forkert (reaktion i realtid). Vi kan udbrede vores kontrol fra ”få i en stikprøve” til ”mange/alle”. Vi kan basere vores kontrol på store datamængder som et menneske ikke ville have kunne overskuet Give anbefalinger: Vi vil kunne hjælpe brugerne på f.eks. Virk.dk med hvilke løsninger de burde være opmærksom på. Brugeroplevelsen vil også kunne gøres mere målrettet og afhængig af om du f.eks. er en lille virksomhed, eller om du er økonomimedarbejder i et stort selskab. Maskinen finder mønstre og hjælpe dig hurtigere frem til relevante indberetningsløsninger på virk eller hjælp på ”startvækst”. Vi vil (hvis man måtte ønske det) også kunne give anbefalinger til virksomhederne af typen: ”Her er de ti brancher hvor man tjente flest penge pr. medarbejder eller pr kapitalandel sidste år” ”Her er det sted i landet hvor bilforhandlere/cafeer/farvehandlere etc. tjente flest penge sidste år. Forbedring af vores policy udvikling med ML-skabt indsigt: Machine learning hjælper med at skabe nye data på ryggen af gamle data. Det kan ske i store mængder. Maskinen kan f.eks. give struktureret viden om den økonomiske situation i en given region ved at se på alle årsrapporter. Maskinen kan også udlede viden af store tekstmængder, så vi ved hvor mange virksomheder der bliver berørt af en ændring krav til opgørelsen af kapitalandele i datterselskaber. Tidligere krævede række sådanne aktiviteter ofte langsommelige og dyre konsulentrapporter. Hvis vi gennemfører AER og kombinerer det med ML vil vi endvidere kunne få tal for den økonomiske udvikling i noget nær realtid
  4. Det vi spørger maskinen om er: Hvad kan vi antage om en virksomhed eller personkredses intentioner, baseret på hvordan de hidtil har opført sig? Eller sagt med andre ord: hvilke spor i eksisterede data om en eller flere personer, giver den stærkest indikation på, at de vil begå moms eller afgiftssvindel i fremtiden. Her ses en typisk virksomhedskontruktion. En personkreds ejer og leder et holdingselskab og en eller flere virksomheder. Ved hjælp af ML kan vi opnå viden om virksomheden og personkredsens tidligere adfærd. Her kan vi f.eks. ”læse” ud af regnskabet v.hj.a. ML, at revisor udtaler at virksomhederne overtræder Moms-, bogføring- og Skattelovgivningen. Vi vil også kunne se om personer f.eks. er tidligere har overtrådt reglerne Det store spørgsmål er nu: Hvad kan vi forvente når de overtager en anden virksomhed? Opgaven for ERST bliver at træne maskinen til a se denne slags situationer og ved hjælp af mønstre at kunne se om der er behov for at sætte virksomheden på ventehylde og aflægge dem et besøg eller kræve yderligere dokumentation inden registreringen kan godkendes, eller om Skat skal underrettes om at denne nye virksomhed er genstand for undring.
  5. Hjælpe og vejlede: Maskinen kan f.eks. i årsrapporter læse anvendt regnskabspraksis, som er fritekst, og give brugeren en ”advis” hvis der ikke er overensstemmelse mellem på den ene side tallene i regnskabet (de mangler eller er forkerte) og på den anden side, den selvanførte regnskabspraksis. Forbedre og skalere kontrol og tilsyn: Vi vil kunne reagere allerede når en brugere forsøger at indsende noget forkert (reaktion i realtid). Vi kan udbrede vores kontrol fra ”få i en stikprøve” til ”mange/alle”. Vi kan basere vores kontrol på store datamængder som et menneske ikke ville have kunne overskuet Give anbefalinger: Vi vil kunne hjælpe brugerne på f.eks. Virk.dk med hvilke løsninger de burde være opmærksom på. Brugeroplevelsen vil også kunne gøres mere målrettet og afhængig af om du f.eks. er en lille virksomhed, eller om du er økonomimedarbejder i et stort selskab. Maskinen finder mønstre og hjælpe dig hurtigere frem til relevante indberetningsløsninger på virk eller hjælp på ”startvækst”. Vi vil (hvis man måtte ønske det) også kunne give anbefalinger til virksomhederne af typen: ”Her er de ti brancher hvor man tjente flest penge pr. medarbejder eller pr kapitalandel sidste år” ”Her er det sted i landet hvor bilforhandlere/cafeer/farvehandlere etc. tjente flest penge sidste år. Forbedring af vores policy udvikling med ML-skabt indsigt: Machine learning hjælper med at skabe nye data på ryggen af gamle data. Det kan ske i store mængder. Maskinen kan f.eks. give struktureret viden om den økonomiske situation i en given region ved at se på alle årsrapporter. Maskinen kan også udlede viden af store tekstmængder, så vi ved hvor mange virksomheder der bliver berørt af en ændring krav til opgørelsen af kapitalandele i datterselskaber. Tidligere krævede række sådanne aktiviteter ofte langsommelige og dyre konsulentrapporter. Hvis vi gennemfører AER og kombinerer det med ML vil vi endvidere kunne få tal for den økonomiske udvikling i noget nær realtid
  6. Identitet At informationen fra identitetspapiret stemmer overens med de indtastede oplysninger om personen på registreringen (MRZ(Machine-Readable-Code)) At identitetspapiret er gyldigt på registreringstidspunktet Spin-off I: identitetspapirer bidrager med præcis metadata om kønsfordelingen i ledelser og bestyrelser Spin-off II: vi kan fremfinde personer registreret med flere enhedsnumre Aktiver, kontrol af vurderingsberetninger I 2018 blev der indberettet 3.554 vurderingsberetninger for selskaber. En tidligere gennemgang af PwC har vist, at 63,5 % af alle vurderingsberetninger indberettet i 2017 er fejlbehæftet. Modellen vil være med til at sikre, at de værdier, som indskydes i selskaber, er reelle. Da modellen vil slå ned i meget specifikke dele af en vurderingsberetning, vil det gøre sagsbehandlingen kortere og nemmere. Årsrapport, revisorerne hjælper Årsrapporter indeholder revisors kommentarer om overtrædelser af love og regler A-lån, Bogføringsloven, Moms og afgifter, Aktivitet ved kapitaltab mangler, A-skat/AM-bidrag 1.st line Kontrol af virksomhedsregistrering på basis af analyse af aktørnetværk, tidligere virksomheders livsforløb, SKAT data. Model i beta, ej produktion 2019.
  7. Erhvervsstyrelsens it-arkitektur er bygget op omkring genbrug af services og fælleskomponenter. Styrelsens hjemmel(§L149) til at anvende andre myndigheders data til kontrol af virksomheder stiller særlige krav til forståelse af data, da der arbejdes med begreber udenfor eget ressort. ERST ML dataplatform er bygget op omkring sporbarhed, forklarlighed og i respekt for det data etiske ansvar som styrelsen har. Fordi en høj etisk standard dikterer sporbarhed, giver dette en positiv sideeffekt ift. evaluering af modellernes præcision og måling af forretningsværdi. Styrelsen arbejder med ud fra et 360 graders forståelse af danske virksomheder som kombinerer den specialiserede indsigt fra maskinlæring, med et kontekstforståelse fra grafteknologi for hvilke mønstre som er udslagsgivende for virksomheders livsforløb. Teknologisk har dette betydet udvidelse af styrelsens infrastruktur med cloud-løsning, containerteknologi til indkapsling af specialiseret teknologi, grafteknologi datastruktur, hændelsesdrevet arkitektur så vi kan reagere i nær-realtid, samt udvidet data governance for sporbarhed og forklarbarhed.
  8. Streaming af data og GDPR Vi ved ikke hvem der stifter selskab før de logger på. Omvendt ønsker vi heller ikke score alle danskere. Derfor er centralt at vi kan samle alle data og anvende dem i det øjeblik som borgeren henvender sig. Vi skal med andre ord ”streame” meget store datamængder på kort tid, da vi kun ønsker at se på de personer og virksomheder der ønsker at oprette og ændre virksomheder uden at vi ”gemmer” dem vi mistænker for at være svindlere i et register. Derved kommer vi uden om en masse GDPR problemstillinger. Machine learning som disciplin Det er en svær ML øvelse, som kræver specialister og grundig forberedelse. Fx er det særlig svært at holde revisionssporet mellem beslutninger taget af maskinen og datagrundlaget. Fordi det skal kunne forklares hvordan vi er kommet fra datakilden til at maskinen er nået frem til dens anbefaling. At reagere i real-time Teknisk er det en svær øvelse at kunne reagere i real time. Dette kræver ny teknologi og vi har måtte flytte dele af vores infrastruktur i skyen for at kunne scalere. Forretninganvendelse af mønstergenkendelse Det stiller store krav til forretningshåndtering, og vi skal have fuldstændig styr på at modeller ikke ”stikker” af fra os. Forretningen skal derfor løbende holde øje med modellerne. Størrelsen af ”netmaskerne” Endeligt er det centralt at huske at ML anvender statistik til at underbygge forretningen. Ved at skrue på modellernes ”confidence” kan vi så at sige ændre netstørrelserne, så vi fx primært går efter de store fisk, og der hvor modellerne er mest sikre.
  9. Data og datas livsforløb: Det er vigtigt, at de beslutninger som Machine learning tager kan genskabes og forklares, samt at beslutningerne er ensartede. Vi vil gradvist få flere og flere data som skabes af machine learning modeller og som indgår i andre modeller som input. Når mange faktorer påvirker data bliver det meget vigtigt og megasvært at forstå og forklare modellernes beslutninger. Med andre ord vi skal udøve god forvaltningsskik, så vi kan forklare ”hvorfor blev virksomhed X” udtaget til nærmere kontrol.