Fraud	Prevention	
A	practical	example	of	graph	databases	in	action
David	Montag	
david@neotechnology.com
A Graph Is Connected Data
Neo4j solves challenges for some of the
most powerful companies in the world
Adidas uses Neo4j to combine
content and product data into a
single, searchable graph database
which is used to create a
personalized customer experience
“We have many different silos, many
different data domains, and in order to
make sense out of our data, we
needed to bring those together and
make them useful for us,” 

– Sokratis Kartelias, Adidas
eBay Now Tackles eCommerce
Delivery Service Routing with Neo4j
“We needed to rebuild when growth
and new features made our slowest
query longer than our fastest delivery -
15 minutes! Neo4j gave us best
solution” 

– Volker Pacher, eBay
Walmart uses Neo4j to give customer
best web experience through
relevant and personal
recommendations
“As the current market leader in graph
databases, and with enterprise
features for scalability and availability,
Neo4j is the right choice to meet our
demands”. 

- Marcos Vada, Walmart
This shift is happening!
Why?
Operational
Databases
Applications
X
Pre-computed Queries
(Data Warehouse / RDBMS)
Real-time & Dynamic Queries

(Graph Database)
Why Graphs Now?
We need something that is:
• Flexible to change
• Scalable to many problems
• Intuitive to understand
• Instantly responsive
Main
Strengths
Main
Risks
• Widespread competence and familiarity
• Integration with wide range of tools
• “Analytical questions in transactional time”
• Flexible and scalable to future needs
• New tools require new competence
• Misunderstandings can result in tool
being wrongly applied
• Incapable of meeting flexibility and
performance demands
• Lacks game changer success stories —
status quo is risky in disruptive times
Ways of Thinking About Data
Pre-computed Queries
(Data Warehouse / RDBMS)
Real-time & Dynamic Queries

(Graph Database)
A Graph Example:
Fraud Prevention
Fraud	Detection	&	Prevention
Types	of	Fraud	
• Retail	Banking	Fraud	
• Insurance	Fraud	
Identity	types	
• Stolen	
• Fake	
• Synthetic	
Types	of	Analysis	
• Discrete	
• Connected
Retail	Banking	First-Party	Fraud
Opening	many	lines	of	credit	with	no	intention	of	paying	them	back
• Tens	of	billions	of	dollars	lost	every	year	by

U.S.	Banks.(1)	
• 25%	of	total	consumer	credit	charge-offs	in	the	
United	States.(2)	
• 10%	to	20%	of	unsecured	bad	debt	at	leading	

U.S.	and	European	banks	is	misclassified,	and	

is	actually	first-party	fraud.(3)
First-Party	Fraud	Impact
(1)		Experian:	 http://www.experian.com/assets/decision-analytics/white-papers/first-partyfraud-wp.pdf	
(2)		Experian:	 http://www.experian.com/assets/decision-analytics/white-papers/first-partyfraud-wp.pdf	
(3)		Business	Insider:	http://www.businessinsider.com/how-to-use-social-networks-in-the-fight-against-first-party-fraud-2011-3
Three	Kinds	of	Identities,	Fraud	Rings
145	Hickory	Rd

Pasadena,	CA
415	Hickory	St

Pasadena,	CA
626-407-1234
626-814-6532
Gartner’s	Layered	Fraud	Prevention	Approach	(4)
(4)	http://www.gartner.com/newsroom/id/1695014
Traditional	Fraud	Prevention
Analysis		
of	users	

and	their	
endpoints
Analysis	of

navigation	
behavior	and	
suspect	patterns
Analysis	of	
anomaly	
behavior	by	
channel
Analysis	of	
anomaly	
behavior	
correlated	across	
channels
Analysis	of	
relationships		
to	detect	
organized	crime	
and	collusion
Layer	1
Endpoint-

Centric
Navigation-

Centric
Account-

Centric
Cross-

Channel
Entity	

Linking
Layer	2 Layer	3 Layer	4 Layer	5
	 DISCRETE	DATA	ANALYSIS	 CONNECTED	ANALYSIS
Pros

Simple

Stops	rookies
Discrete	Data	Analysis
Revolving

Debt
INVESTIGATE
INVESTIGATE
Number	of	accounts
Cons

False	positives

False	negatives
Connected	Analysis
Revolving

Debt
Number	of	accounts
PROS

Detect	fraud	rings

Fewer	false	negatives
Value

Effective	in	detecting	some	of	the	
most	impactful	attacks,	even	from	
organized	rings	
Challenge

Extremely	difficult	with	traditional	
technologies
For	example	a	ten-person	fraud	bust-out	is	$1.5M,	assuming	100	false	identities	

and	3	financial	instruments	per	identity,	each	with	a	$5K	credit	limit
Connected	Analysis	with	Neo4j
Demo
Insurance	Fraud					”Whiplash	for	Cash”
Paper	Collisions	
Insurance	scammers	invent	automobile

accidents	complete	with	fake	drivers,

passengers	and	witnesses
Whiplash	for	Cash				Example
Accidents
Cars
Doctor Attorney
People
Drives
Is	Passenger
Drivers

Passengers

Witnesses
View	of	fraud	ring	

in	a	graph	database
Modeling	Insurance	Fraud	as	a	Graph
Accident

1
Accident

2
Person

1
Person

2
Person

3
Person

4
Person

5
Person

6
Car

1
Car

2
Car

3
Car

4
INVOLVES
DRIVES
REPRESENTS
WITNESSES
ADJUSTS
HEALS
Doing	Connected	Analysis	is	Challenging
• Large	amounts	of	data	and	relationships	
must	be	processed	
• New	data	and	relationships	are	continually	
being	added	
• Fraud	rings	must	be	uncovered	in	

real-time	to	prevent	fraud
Adopting a Pattern-oriented Mindset
Search-oriented
• Good when you know
exactly what you’re
looking for
• Primarily based on explicit
search criteria
Pattern-oriented
• Good when you want to
suggest what might fit
• Primarily based on implicit
information, often many
“hops” away
The Case For Innovation with Graphs
During past 20 years, society has
become hyperconnected.
We considered how regular people
tend to think and reason, and modeled
Neo4j to match that.
Neo4j allows you to naturally map
together the data that matters to you in
a graph — like a mind map!
…
Graph structure scales to many
problems, and is highly flexible to
change.
…
Unlock the business value of
connections and relationships in data.
…
How can we discover insights
hidden behind the complexity?
…
Traditional technology does not
handle complexity well.
…
Information-wise, gone from a small
town to a metropolis. Complexity has
exploded.
…
Need for better insights driven by
competition and disruption. There is a
shift happening now.
The world’s largest companies rely
on Neo4j. The competitive
advantage is real.
…
In uncertain times, many consider
the risks of change, but what are
the risks of not adapting?
…
Get in the driver’s seat. Be the
bringer of innovation.
…
The Problem The Solution The Future
THANK YOU!
(me)-[:ASKS_FOR]->(tweet {say: “Neo4j GraphDay:
what a time to be alive!”})-[:FROM]->(you)

GraphDay Stockholm - Fraud Prevention