Transcript EY - Advanced data analytics in investigations 2015

Advanced Data Analytics in
Fraud Identification
February 23, 2015
Agenda
►
Current challenges
►
Data analytics defined
►
What are clients saying?
►
Technology
►
Case Examples
►
Questions
1
Current challenges err…opportunities
Tax evasion
Off shore
Geopolitical
Government contracts
Conflict
Labor relations
Customer complaints
minerals
Transparency
Reputation
Suppliers
Politically Exposed Persons
Bribes
Investigations
Corruption
Cyber Sanctions
Litigation
Self certification
2
HSE
Fines
United States cases
US Pending investigations
►
As of July 2014, 106 publicly disclosed investigations pending
ABM Industries Incorporated
Accenture PLC
Agilent Technologies Inc..
Airbus Group
Alstom SA
Delphi Automotive PLC
Deutsche Bank AG
Deutsche Post AG (DHL)
DreamWorks Animation SKG Inc..
Dun & Bradstreet Corporation
Key Energy Services Inc..
Kimco Realty Corporation
KKR & Company LP
Las Vegas Sands Corp
Layne Christensen Company
Analogic Corporation
Eli Lilly and Company
Mead Johnson Nutrition Company
Anheuser-Busch InBev SA/NV
AstraZeneca PLC
Embraer SA
Ericsson AB
Merck & Co Inc..
Microsoft Corporation
Mondelēz International Inc. (formerly Kraft Foods Inc.)
Avon Products Inc.
Expro International Group PLC
Barclays PLC
Beam Inc.
BHP Billiton Ltd
Bio-Rad Laboratories Inc.
Blackstone Group LP
Bristol-Myers Squibb Company
Brookfield Asset Management Inc.
FedEx Corporation
Freeport-McMoRan Copper & Gold Inc.
Fresenius Medical Care AG & Co KGaA
General Cable Corporation
GlaxoSmithKline PLC
Gold Fields Limited
Goldman Sachs Group Inc.
Morgan Stanley
Motorola Solutions Inc.
MTS Systems Corporation
National Geographic
NCR Corporation
Net 1 UEPS Technologies Inc.
News Corporation
Bruker Corporation
Goodyear Tire and Rubber Company
Grifols SA (Talecris Biotherapeutics Holdings Corp)
Holdings Corp)
Halliburton Company
Nordion Inc.
Harris Corporation
Hyperdynamics Corporation
Image Sensing Systems Inc.
Ingersoll-Rand PLC
International Business Machines Corporation
Corporation
Johnson Controls Inc.
Olympus Corp
Oracle Corporation
Orthofix International NV
Owens-Illinois Group Inc.
UBS AG
Universal Entertainment Corp
Universal Music Group (Vivendi)
(Vivendi)
Viacom (Paramount Pictures)
VimpelCom Ltd
Wal-mart Stores Inc.
Panasonic Corporation
PTC Inc.
Walt Disney Company
WS Atkins PLC (PBSJ Corp)
JPMorgan Chase & Co
Juniper Networks
Park-Ohio Industries Inc.
Protective Products of America Inc.
BSG Resources Ltd
Central European Distribution Corporation
Chestnut Consulting Inc.
Cisco Systems Inc.
Citigroup Inc.
Credit Suisse Group AG
Cobalt International Energy Inc.
Comcast (NBCUniversal Inc.)
Cubist Pharmaceuticals Inc. (Optimer Pharmaceuticals Inc.)
Dialogic Inc.
3
Novartis AG
Och-Ziff Capital Management Group LLC
Qualcomm Incorporated
Quanta Services Inc..
Rolls Royce PLC
Sanofi SA
SBM Offshore NV
Sciclone Pharmaceuticals Inc..
Sensata Technologies Holding NV
NV
Siemens AG
SL Industries Inc.
Smith & Wesson Holding Corporation
Corporation
Société Générale SA
Sony Corporation
STR Holdings Inc.
Tata Communications Limited
Tesco Corporation
TeliaSonera AB
Teva Pharmaceutical Industries Limited
Industries Limited
Forensic Data Analytics Defined
4
Forensic Data Analytics
Analytics Defined
Forensic Data Analytics (“FDA”): Refers to the multi-disciplinary service for the efficient and
cost effective identification of relevant information in large-scale client datasets for a wide range of
activity including the analysis of data to obtain meaningful insights for investigative, legal,
regulatory, anti-fraud or risk mitigation matters.
FDA combines the extensive use of data, statistical and quantitative analysis, and explanatory and
predictive models to guide and identify issues and areas warranting further review.
Outputs from FDA can allow companies to generate legally defensible fact-based evidence in
order to drive decisions and focus investigative efforts where they matter, with the aim of achieving
favorable outcomes.
5
FDA supports the Corporate Integrity & Compliance Framework
Risks
Detect
Corporate Integrity &
Compliance
Framework
Respond
Protect
Forensic Data Analytics
Reactive
Proactive
6
How is fraud detected?
Source: ACFE 2010 Report to the Nations On Occupational Fraud
7
48.5% by tip
or accident
Forensic Data Analytics Maturity
Focus
Forensic single version of the truth
“Gather the hay into a haystack”
Errant behaviour detection
“Find the needle”
Ongoing monitoring & protection
“Prevent the needle being lost in the first place”
Capabilities
(in order of maturity)
1. Forensic data discovery & extraction
2. Data joining and filtering
3. Application of rules
4. Matching to external data (e.g. sanctions lists)
sanctions lists)
5. Unstructured text mining (e.g. key word search)
7. Visualisation and drill-down
8. Case management and feed-back look
8
Higher
Detection
Rate
Lower
False
Positive
Rate
Understanding the FDA approach
Database Management Tools
unstructured
structured
TRANSACTIONAL
DATA
MASTER &
REFERENCE
DATA
BUSINESS
INTELLIGENCE
DATA
SOCIAL MEDIA
DATA
9
Understanding the FDA approach
Database Management Tools
Detection
Tools
unstructured
structured
TRANSACTIONAL
DATA
Rules-based tests
Text mining &
search
MASTER &
REFERENCE
DATA
BUSINESS
INTELLIGENCE
DATA
SOCIAL MEDIA
DATA
Big data and/or SQL
server data processing platform
VISUALIZATION
& RISK RANKING
10
Understanding the FDA approach
Database Management Tools
Detection
Tools
unstructured
structured
Investigation
Tools
Pattern Matching
TRANSACTIONAL
DATA
Rules-based tests
Text mining &
search
MASTER &
REFERENCE
DATA
Statistical anomalies
& Predictive
BUSINESS
INTELLIGENCE
DATA
SOCIAL MEDIA
DATA
Big data and/or SQL
server data processing platform
VISUALIZATION
& RISK RANKING
11
Case Manager, Task
Delegation and
Data Refresh /
Scripting
Automation
Understanding the FDA approach
Database Management Tools
Detection
Tools
unstructured
structured
Investigation
Tools
Pattern Matching
TRANSACTIONAL
DATA
Rules-based tests
Text mining &
search
MASTER &
REFERENCE
DATA
Statistical
& Predictive
BUSINESS
INTELLIGENCE
DATA
SOCIAL MEDIA
DATA
Case Manager, Task
Delegation and
Data Refresh /
Scripting
Automation
Big data and/or SQL
server data processing platform
VISUALIZATION
& RISK RANKING
Case
Review
12
Understanding the FDA approach
Database Management Tools
Detection
Tools
unstructured
structured
Investigation
Tools
Pattern Matching
TRANSACTIONAL
DATA
Rules-based tests
Text mining &
search
MASTER &
REFERENCE
DATA
Statistical
& Predictive
BUSINESS
INTELLIGENCE
DATA
SOCIAL MEDIA
DATA
Case Manager, Task
Delegation and
Data Refresh /
Scripting
Automation
Big data and/or SQL
server data processing platform
VISUALIZATION
& RISK RANKING
Case
Review
Management
Review
Findings &
Recommendations
13
Understanding the FDA approach
Database Management Tools
Detection
Tools
unstructured
structured
Investigation
Tools
Pattern Matching
TRANSACTIONAL
DATA
Rules-based tests
Text mining &
search
MASTER &
REFERENCE
DATA
Statistical
& Predictive
BUSINESS
INTELLIGENCE
DATA
SOCIAL MEDIA
DATA
Transaction review,
Risk ranking, Case
Manager, Data Refresh
/ Scripting
Automation
Big data and/or SQL
server data processing platform
VISUALIZATION
& RISK RANKING
Repeat the process:
Continuous Monitoring
(On-Site, centralized, outsourced)
Case
Review
Management
Review
Findings &
Recommendations
14
What Are Clients Saying?
15
What Are Clients Saying?
FDA Technology
16
What Are Clients Saying?
FDA Challenges
With respect to forensic data analytics, what would you say is your single
biggest challenge or requirement in your organization?
Getting the right tools or expertise for FDA
26%
Challenges with combining data across various IT systems
15%
Improving the quality of the analysis process
15%
Convincing senior management or the company about the benefits of
FDA
10%
FDA is too expensive
10%
To prevent fraud rather than discover fraud
9%
Poor quality or lack of accuracy in the data
8%
Difficulty in adapting FDA to comply with different regulations in
various markets
6%
Spreading the FDA culture across different Business Units
6%
Lack of human resources or manpower to operate FDA
5%
To identify fraudulent information across large data sets
5%
Huge volume of data to analyze
4%
FDA is not prevalent to the culture
3%
FDA producing positive results to indicate and prove any fraud or
bribery that is occurring
3%
Uncertainty about the relevance of FDA in the Company
2%
0%
5%
17
10%
15%
20%
25%
30%
Technology
18
Technology
Text Analytics
►
Keyword library
►
Concept analysis
►
Entity extraction
►
Text is everywhere
19
Document/data review analytics
►
►
►
Concept induction and linguistic analysis
►
Keywords and Ontologies
►
Emotive tone and ethical issue detection
►
Topic modeling, concept mining, entity extraction
Social network and actor analysis
►
Centrality and proximity
►
External domains, family and friends
Ontologies
20
Ontologies
►
Keywords alone are not as
effective to reliably identify
key concepts
►
Ontologies allow us to
capture concepts
appearing in unstructured
textual data
►
Can be developed in
automated or manual
ways, and are reusable
across engagements
21
Stock ontologies
►Non-Responsive
(1,832 classifiers; 19,188 terms)
►NR
Business
►Resumes, “doughnuts in the kitchen,”
newsletters, itineraries
►NR Junk
►Spam, fantasy football, baby pictures
►Emotive Tone (66 classifiers; 4,101 terms)
►Angry, Confused, Secretive, Surprised
►Ethical Issues (419 classifiers; 5,042 terms)
►Discrimination, workplace safety, price fixing,
personal problems
22
Technology
Risk Scoring
Review breaches on targeted
analytics
23
Filter by selected analytics
Technology
Geocoding and Heat Maps
Identify global epicenters of activity, as well as anomalies
Hotspots of
activity are
easily
identified
24
Technology
Communication Analytics
WH
O`
Social Networking
Who is talking to whom?
WHA
T
WHE
N
Concept Clustering
about what?
Communication Over Time
over which time period?
WHY
Sentiment Analysis
how do they feel?
• People-to-people analysis
• Top words mentioned
• When communications occur
• Positive vs. Negative Sentiment
• Entity-to-entity analysis
• Key concepts / topics
• Top 10 negative journal entries
• Map communication lines
to organization chart
• Top or unusual dollar amounts
• Communication spikes
around key business events
• Sensitive words / phrases
• Top 10 angry emails
• Top 10 most concerned emails
• Customer survey analysis
• Employee survey analysis
25
Technology
Email Analytics
► Identify
the strength of known relationships
► Identify
unknown relationships
► Can
be performed on email logs, don’t need
actual emails
26
Entity analytics
Unusual connection
between entities
27
Technology
Social Media Analysis
28
Email Analytics
►
Research example: FCPA bribery case
Keyword hits as a percentage of total emails
Incentive/
pressure terms
Opportunity
terms
Rationalization
terms
Investigation timeframe, September to March
29
Convergent analytics model
Anti-fraud library
of journal entry
and cash
disbursement
tests
Text analytics:
“who,” “what,”
“when,” “where”
Suspicious
transaction
profiling and
predictive
modeling
Vendor master
Employee master
Travel & Entertainment
Accounts Payable
General Ledger
Accelerated
decision
support and
dynamic
reporting
EY/ACFE keyword
library of misappropriation &
bribery/
corruption terms
(multi-language)
30
Vendor
background
checks and
politically
exposed
persons
Emotive tone -- Secretive
From: Donna
Sent: Wednesday, November 9, 2011 10:45 AM
To: Nikki
Subject: RE: Shhhhh
Absolutely.
-----Original Message----From: Nikki
Sent: Wednesday, November 09, 2011 10:45 AM
To: Donna
Subject: Shhhhh
Please don't mention our convo to anyone! I am high risk so I want to be sure that everything is safe
and where it should be.
31
Thank You
Scott Clary
Ernst & Young LLP
Principal
Fraud Investigation & Dispute Services
Houston, TX
(713) 750-1593
[email protected]
32