semantic applications for financial services

29
Semantic Applications for Financial Services David Newman Strategic Planning Manager Enterprise Technology Architecture and Planning Wells Fargo Bank June 23, 2010

Upload: davidsnewman

Post on 25-Jan-2015

2.224 views

Category:

Technology


1 download

DESCRIPTION

This presentation provides an overview of the business and technical drivers for building financial service applications using Semantic Technology. Multiple use cases are provided as examples.

TRANSCRIPT

Page 1: Semantic Applications for Financial Services

Semantic Applications for Financial Services

David NewmanStrategic Planning ManagerEnterprise Technology Architecture and PlanningWells Fargo Bank

June 23, 2010

Page 2: Semantic Applications for Financial Services

June 23, 2010

Disclaimer

2

The content in this presentation represents only the views of the presenter and does not represent or imply acknowledged adoption by Wells Fargo Bank. Examples used within are purely hypothetical and are used for illustrative purposes only and are not intended to reflect Wells Fargo policy or intellectual property.

Page 3: Semantic Applications for Financial Services

June 23, 2010

What Benefits Does Semantic Technology Provide for Financial Services Organizations?

What are some of the business and technology drivers for Semantic Technologies from a Financial Services perspective? What are some of the critical business and technology

problems that Semantic Technology attempts to remedy?

What are some limitations with conventional Information Technologies that Semantic Technology improves upon?

What are some Financial Service use cases that can demonstrate benefit by using Semantic Technology?

3

Page 4: Semantic Applications for Financial Services

June 23, 2010

IT Organizations are often asked by the Business to: provide a holistic, comprehensive, integrated view of the

Customer

fulfill major data and system integration initiatives

cross organizational and system boundaries to accomplish this

deliver all of the above functionality faster, cheaper, smarter

4

Common IT Challenges at Financial Services Firms …

Page 5: Semantic Applications for Financial Services

June 23, 2010

This must often be accomplished in environments where there exists: a preponderance of incompatible data definitions, vocabulary

multiple incompatible physical data and file formats, databases, storage mechanisms

a proliferation of fragmented, redundant data

a proliferation of unstructured data that is inaccessible to most users

dissonance between the business stakeholders definition of data and processing rules and how such data and rules are actually codified within application software

Can result in high costs, slipped dates

5

Often Requires IT to Surmount Difficult Obstacles …

Page 6: Semantic Applications for Financial Services

June 23, 2010

Requires New and Innovative Tools that will help IT organizations to: standardize and unify the meaning of data across the enterprise

capture and persist business and technical knowledge as information assets

foster data integration despite organizational boundaries

give greater control to the Business for definitions of data and business rules

produce better results faster and cheaper than conventional technologies

Semantic Technology can help to achieve these goals!

6

That May Not Always Be Solved by Conventional Technologies

Page 7: Semantic Applications for Financial Services

June 23, 2010 7

Conventional Information Technologies: What’s Wrong?

Data Schema

New Data Entity

Physical Database

New Physical Table for New Entity

Application Software

Business Rules in Code

Access

Update

Define

Knowledge is encapsulated in opaque software

Data organization is tightly coupled with the schema

Data schemas reflect limited knowledge

Schemas enforce limited data integrity

Data is siloed as is its meaning

Conventional Technology Data Definition and Access Patterns

Data fragmentation

Data redundancy

Data incompatibility

Labor intensive tasks

High costs

Slow time to market

Page 8: Semantic Applications for Financial Services

June 23, 2010

Conventional Relational Database Example

8

CustomerCustId EntityType TaxId

David_Newman Person 999-99-9999

Checking AccountAcctId Balance Overdrafts

0001-12345 $1,000 0

Customer Account Relationship

CustId AcctId ProductId

David_Newman 0001-12345 DDA

Product Hierarchy

ProductID Group Domain

DDA DEP CONS

PersonCustId Last Name First Name

David_Newman Newman David

Product Catalog

ProductID Name

DDA Deposit Account

Awareness of the physical organization of data is necessary

Many tables are often required to capture entities and their relationships

Entity relationships are realized by joining data, mainly by its keys

Guided by Closed World Assumption – if data is not present it does not exist

Page 9: Semantic Applications for Financial Services

June 23, 2010 9

What is Semantic Technology? Major step towards reducing data chaos

Based upon Description Logic A mathematically verifiable symbolic

logic that allows reasoning about entities and the many properties that describe entity relationships

Describes entities in terms of: Concepts (classes)

Relationships (properties)

Individuals (instances)

Makes inferencing possible Infers relationships and memberships in

classes per axioms via a “Reasoner”

Guided by Open World Assumption If data is not present it may still exist!

Subject(domain)

Subject(domain)

Predicate (property)

Predicate (property)

Object(range)

Object(range)

RDF Triples/ Statements

Aligns linguistically with how we think and speak!

Jackson Pollock “Convergence”

Page 10: Semantic Applications for Financial Services

June 23, 2010 10

Semantic Technology: How does it Help?New Data Entity

Ontology / Semantic SchemaPhysical Database

Some Business Rules Added to Ontology

Application Software

Some Inferred Data

Some Business Rules Removed from Code

Physical Format Unchanged after New Data Entity Added

Access

Update

Define

Semantic Technology Data Definition and Access Patterns Knowledge is open and represented by an

ontology Meaning and relationships of data defined Data organization is decoupled from the

schema Inferencing creates new knowledge All semantic data is Web addressable

TBox (terminology)

ABox (assertions)

Meaning is consistent

Knowledge is accessible

Applying rules to data is easier and less costly

Data access costs should be lower

Faster time to market

TBox

Page 11: Semantic Applications for Financial Services

June 23, 2010

Product

Consumer Product

Consumer Credit

Retail Deposit

Credit Card

HomeEquity

RetailChecking

SavingsDeposit

Customer

Financial Information Ontology

11

Account Status

ClosedOpen

Business Partner

Person

BusinessEntity

Country

United States

Credit Card Eligible Customer

Credit Risk Retail Customer

Fraud Risk Retail Customer

Gold Credit Card Eligible Customer

Event Type

Online Login

AccountOpen

ChangeAddress

TransferFunds

accountStatus

Event

Online Login Event

Suspicious Online Login Event

hasEvent eventForCustomer

isEligibleFor

hasPre-Qualified

hasAccounttitleHolderhasIdentity

is Customer

hasEventType

productType

isAccount

onlineLoginEventLocation

eventForCountry

describes Event

accountForStatus

Account

ConsumerAccount

DepositAccount

Checking Account

Consumer Credit

Account

HELOC

Bad Country

Bad Country X

Page 12: Semantic Applications for Financial Services

June 23, 2010

Semantic RDF “Triple Store” Example

12

Subject Predicate ObjectCustomer rdf:type owl:Class

Customer rdfs:subClassOf Thing

CheckingAccount rdf:type owl:Class

CheckingAccount rdfs:subClassOf DepositAccount

hasAccount rdf:type rdf:Property

hasAccount rdfs:domain Customer

hasAccount rdfs:range Account

hasAccount owl:inverseOf titleHolder

titleHolder rdf:type rdf:Property

titleHolder rdfs:domain Account

titleHolder rdfs:range Customer

titleHolder owl:inverseOf hasAccount

David_Newman rdf:type Customer

0001-12345 rdf:type CheckingAccount

David_Newman hasAccount 0001-12345

0001-12345 titleHolder David_Newman

0001-12345 rdf:type DepositAccount

InferredTriples

ABox

Assertions FactsData

TBox

TerminologyOntology Schema

Every element is a Web addressable URI!

Page 13: Semantic Applications for Financial Services

June 23, 2010

How Can We Apply Semantic Technology to Specific Financial Services Use Cases for Maximum Benefit? The following use cases represent a sampling of ways that

Semantic Technology can be effectively applied in a Financial Services organization:

1. Linked Enterprise Data: 360 Degree Customer View

2. RDFa Enablement of Online Financial Services and Products

3. Fraud Detection

4. Eligibility and Suitability Rules

5. Credit Risk Management

6. Integrated Financial Statements

7. Concept Extraction and Categorization from Unstructured Text

8. Market Intelligence for Investment Analytics

13

Page 14: Semantic Applications for Financial Services

June 23, 2010

Linked Enterprise Data: 360 Degree Customer View

14

Semantically enabled data that is Web addressable and “inter-linked” across the enterprise

Transcends organizational boundaries and provides universal access to data wherever it resides within the enterprise (and externally)

Reduces redundancy by obtaining data from its “virtualized” source

Integrating data in a siloed environment is a major win for Financial Information Systems

Deposits

Stores

OnlineBanking

Loans

Corporate

Customer

“Customer Centric”EnterpriseData Cloud

360 degree view of customer

“KYC”Know YourCustomer

Twitter

Facebook

Credit Bureau

Risk

Page 15: Semantic Applications for Financial Services

June 23, 2010

Financial Ontologies

RDFa Enablement of Online Financial Services & Products

15

Semantic search engine optimization, semantic marketing and sales

Growing evidence that RDFa will: improve rank on Search Engines

increase traffic to site

improve click-thru rates

FIs can RDFa enable: products information , e.g. terms, rates for:

loans, CDs, checking, savings, etc.

services e.g. bill payments, financial advice

Context based semantic search

Semantic agents Agent initiated search

Agent initiated filtering

Agent initiated transactions

Base Ontology

RDFa Enabled Web Page

Page 16: Semantic Applications for Financial Services

June 23, 2010

Fraud Detection

Effective Pattern Detection and Event Correlation From a set of known facts (Events), and a set of rules (Axioms), a

Reasoner infers membership in classes that reflects a specific pattern indicating fraud risk

Expectation of reduced cost in comparison to conventional technologies

alter Tbox dynamically to define new patterns and relationships

yields faster results, lower maintenance and deployment costs

Utilizes Open World Assumption (OWA) Reasoning

Effective Link Analysis Once a common entity is known e.g. bad phone number; other risky

relationships may emerge by invoking queries that perform Graph Pattern Matching to identify fraud networks or other victims

Unique Naming Assumption not supported by OWA

16

Page 17: Semantic Applications for Financial Services

June 23, 2010

Customer

Retail Customer

Fraud Risk Pattern

17

Country

BadCountry

Bad Country X

Product

Consumer Product

Event Type

Online Login AccountOpen

ChangeAddress

hasEvent

hasAccount

isEventType

productType

onlineLoginEventLocation

Account

Consumer Account

Equivalent Class:

Account and productType some ConsumerProduct

Event

Online Login Event

Suspicious Online Login Event

Equivalent Class:

eventForCustomer some Customer and isEventType value OnlineLogin and onlineLoginEventLocation some BadCountry

Which consumer customers might be atrisk of Online Account Takeover Fraud?

Fraud Risk Retail Customer

Equivalent Class:

RetailCustomer and hasEvent some SuspiciousOnlineLoginEventand hasEvent some (Event and isEventType value AccountOpen)and hasEvent some (Event and isEventType value ChangeAddress)

Answers the Query:onWatchListFor some OnlineAccountTakeoverand returns a set of customers at risk

onWatchListFor

Risk Category

OnlineAccountTakeovern

Equivalent Class:

Customer and hasAccount some ConsumerAccount

eventForCustomer

Note: Semantic solutions, other than OWL DL, could also be used to achieve the same results

Page 18: Semantic Applications for Financial Services

June 23, 2010

Eligibility and Suitability Rules

Outbound marketing campaign extractions based upon pre-qualification of customers for specific products

Cross-Sell and Offer Generation

Online preferences and Personalization

Eligibility rules for: Account Acquisitions

Loan Originations

Money Movement Transactions

18

Page 19: Semantic Applications for Financial Services

June 23, 2010

Product

Credit Card Eligibility:

19

Account Status

ClosedOpen

Consumer Product

Consumer Credit

Retail Deposit

ConsumerCredit Card

RetailChecking

SavingsDeposit

Customer

Credit Card Eligible Customer

Account

Retail Checking Account accountStatus

isEligibleFor

hasAccount

productType

Doublebalance

IntegeroverDraftsMos

Equivalent Class:

Customer and hasAccount some (RetailCheckingAccount and accountStatus value Open and balance some double[> 1000.00] and overdraftsMos some nonNegativeInteger[< "1”])

Equivalent Class:

Account and productType some RetailChecking

Answers the Query:isEligibleFor some BasicCreditCardand returns a set of eligible CustomersCan also answer:isEligibleFor some GoldCreditCardand returns a set of Customers eligible for all Premium Consumer Credit Card Products

Which consumer customers are eligible for a Consumer Credit Card?

hasPrequalified

Gold Credit Card Eligible Customer

Equivalent Class:

CreditCardEligibleCustomer and hasAccount some (RetailCheckingAccount and balance some double[> 50000.00]

BasicCreditCard

Gold Credit Card

Page 20: Semantic Applications for Financial Services

June 23, 2010

Credit Risk Management

Identify levels of credit risk by vetting a set of facts collected about the customer with a set of rules that govern risk

20

Page 21: Semantic Applications for Financial Services

June 23, 2010

Credit Risk Management:

21

Customer

Credit Risk Consumer Customer

Account

Retail Checking Account

hasAccount

productType

IntegerdelinquentDays Integer overDraftsPastMonth

Equivalent Class:

Customer and ((hasAccount some (RetailCheckingAccount and (overdraftsPastMonth some nonNegativeInteger[> 1]))) or (hasAccount some (ConsumerCreditAccount and (delinquentDays some nonNegativeInteger[>= 30]))))

Equivalent Class:

Account and productType some RetailChecking

Answers the Query:atRiskFor some ConsumerCreditRiskand returns a set of Customers at risk

Which consumer customers are at risk from a credit perspective?

Product

Consumer Product

Consumer Credit

Retail Deposit

Credit Card

HomeEquity

RetailChecking

SavingsDeposit

Risk Category

ConsumerCreditRisk

Consumer Credit Account

Equivalent Class:

Account and productType some ConsumerCredit

atRiskFor

Page 22: Semantic Applications for Financial Services

June 23, 2010

Integrated Financial Statements Semantically aligned financial statements

ability to roll up financial information across disparate internal business units and external companies so that Financial Reports can be published/understood with higher levels of reliability and trust

attain holistic view of organization’s financial health

improve financial risk management for enterprise and investments

XBRL (Extensible Business Reporting Language) SEC has mandated US public companies file financial reports using

XBRL 3Q09 (proliferating globally)

RDF/OWL enabled XBRL XML Leverage benefits of semantic technology using XBRL

W3C Interest Group

Rhizomik Initiative (ReDeFer XML2RDF,XSD2OWL)

22

Page 23: Semantic Applications for Financial Services

June 23, 2010

Concept Extraction and Categorization from Unstructured Text eDiscovery - categorizing, searching and accessing structured and

unstructured content often for legal purposes Semantic metadata tags associated with content for optimized search

Intentionally asserted by user

Automatically asserted by using semantic entity extraction and Natural Language Processing (NLP) tools to identify conceptual meaning of unstructured content

Customer Related Concept Extraction Semantic parsing of unstructured text using NLP and concept extraction

to identify: Meaning of customer emails for Voice of the Customer

Directing notifications to Bankers based upon accurate categorization of content from text for Lead Management purposes. etc.

23

Page 24: Semantic Applications for Financial Services

June 23, 2010

Market Intelligence for Investment Analytics

Open Source Intelligence capabilities leverage semantic analysis of news feeds and Web content

pertaining to companies of interest for investment purposes

Low Latency Critical Notifications to Analysts provides rapid categorization of content and processing

against a set of semantically defined criteria (axioms) in order to send notifications to analysts for further investigation when an event of interest is identified

24

Page 25: Semantic Applications for Financial Services

June 23, 2010

Better Service Oriented Architecture Service Oriented Architecture

SOA is often foundational to Financial Service architectures Canonical Semantic Data Schema can auto-translate data

content from one interface protocol to another, increasing the level of interoperability

Semantic Service Repository Elements within schema defined as entities within an ontology

ensuring semantic alignment, clarity and expressiveness

Service Registry and Discovery Requires capability to advertise and locate service interfaces

defined by a Service Registry using semantic content for unambiguous context based search

25

SOA

Page 26: Semantic Applications for Financial Services

June 23, 2010 26

Implications for Enterprise Architectureand Data Management Organizations Enterprise Ontology Governance:

Manage and provide standards and quality control for enterprise semantic content across the enterprise

Enable and manage an enterprise Ontology Repository Limit risk of siloed ontologies, enable federation of ontologies Evolve enterprise “Upper Ontology”

Business Semantics Management Encourage and incubate Line of Business Ontologies Influence LOBs to open their data silos for Linked Enterprise Data Provide business friendly interface that front-ends the Ontology

Semantic Technology Governance Ensure effective Access Control and Trust mechanisms are provided Ensure effective Quality of Service mechanisms are provided to achieve

desired performance, availability, recoverability (etc.) standards

EA

Page 27: Semantic Applications for Financial Services

June 23, 2010 27

(Some) Providers of Semantic Technology

Ontology Editors

Triple/RDF Stores

Middleware

PelletRacerPro

Reasoners

Sesame

OWLAPILanguages

NetKernel

Business SemanticsManagement

Page 28: Semantic Applications for Financial Services

June 23, 2010 28

(Some) Recommended Books

Page 29: Semantic Applications for Financial Services

June 23, 2010 29

Follow-up or Questions

Email: [email protected]