HomeWebcastEffective Identity Fraud Prevention and Detection with Machine Learning: Tips and Strategies
Online CLE Identity Fraud Prevention CLE

Effective Identity Fraud Prevention and Detection with Machine Learning: Tips and Strategies

Live Webcast Date: Thursday, August 29, 2019 from 12:00 pm to 2:00 pm (ET)
Business & Corporation CLE & CPELive Webcast
Add to Calendar 08/29/2019 11:00 am 08/29/2019 2:00 pm America/New_York The Knowledge Group Webinar: Effective Identity Fraud Prevention and Detection with Machine Learning: Tips and Strategies The event starts in an hour. If you have not registered, please go to the link below to sign up today.https://www.theknowledgegroup.org/checkout/?add-to-cart=145014\r\n The rise of machine learning technology in recent years has provided businesses, particularly financial institutions, with advanced capabilities in preventing and detecting identity fraud at account opening. However, there are many arguments regarding the best practices in identity fraud defense, including the role of machine learning in pinpointing risks. The message is clear: a strong fraud defense requires machine learning, but equally requires strong underlying data and well-practiced fraud operations procedures to make it effective. \n \nBusinesses should consider a number of factors before moving forward with a machine learning-based fraud detection and prevention solution. \n \nIn this LIVE Webcast, a seasoned panel of thought leaders and professionals brought together by The Knowledge Group will provide and present an in-depth analysis of employing machine learning in detecting and preventing fraud. Speakers will also provide practical tips and strategies to ensure that potentials are maximized, and pitfalls are mitigated. \n \nSome of the major topics that will be covered in this course are: \n \n \n Recent Identity Fraud Threats & Trends at Financial Institutions \n Critical Qualities of Identity Fraud Defense Strategies \n Fraud Prevention and Detection in Machine Learning – Fundamentals \n Practical Best Practices and Strategies in Fraud Defense \n What Lies Ahead \n https://www.theknowledgegroup.org/webcasts/identity-fraud-prevention-machine-learning/

Online CLE Identity Fraud Prevention

Join us for this Knowledge Group Online CLE Identity Fraud Prevention Webinar. The rise of machine learning technology in recent years has provided businesses, particularly financial institutions, with advanced capabilities in preventing and detecting identity fraud at account opening. However, there are many arguments regarding the best practices in identity fraud defense, including the role of machine learning in pinpointing risks. The message is clear: a strong fraud defense requires machine learning, but equally requires strong underlying data and well-practiced fraud operations procedures to make it effective.

Businesses should consider a number of factors before moving forward with a machine learning-based fraud detection and prevention solution.

In this LIVE Webcast, a seasoned panel of thought leaders and professionals brought together by The Knowledge Group will provide and present an in-depth analysis of employing machine learning in detecting and preventing fraud. Speakers will also provide practical tips and strategies to ensure that potentials are maximized, and pitfalls are mitigated.

Some of the major topics that will be covered in this course are:

  • Recent Identity Fraud Threats & Trends at Financial Institutions
  • Critical Qualities of Identity Fraud Defense Strategies
  • Fraud Prevention and Detection in Machine Learning – Fundamentals
  • Practical Best Practices and Strategies in Fraud Defense
  • What Lies Ahead

Agenda

Kevin King, Head of Marketing
ID Analytics, a Symantec Company

Clyde Langley, Vice President, Fraud Risk Management, Financial Crimes Risk Management
Charles Schwab & Co., Inc.
  • Consistent and accurate data feeds from multiple datapoints are the critical first step to implementing ML for fraud prevention and detection
  • ML requires sufficient activity and fraud instances to “learn” normal vs anomalous activity and detect fraud
  • Identify the right datapoints and ingest clean data for the model to learn normal activity
  • Fraud incident rates in the channel need to be sufficient to develop better predictive capabilities to detect fraud through ML
  • Understanding of ML fraud detection implementation decisions
  • ML implementation approach should be based on in-house capabilities (Wayne, this will likely tie-in directly based on your expertise
  • Consider criminals’ ability to identify and bypass your anti-fraud controls in implementing ML
  • Benefits (selling points) of ML in anti-fraud prevention and detection can reduce alerts and minimize operational costs while improving CX and mitigating fraud

Wayne Shoumaker, SVP, Quantitative Analytics Manager
Wells Fargo
  • Identity fraud spans a broad range of threats
    • Fraud models once considered outside of the scope of model risk management (MRM) are now considered in scope
    • Best practices need to align with FRB SR 11-7 and MRM
    • Regulatory guidance for a model’s conceptual soundness, outcomes analysis, and ongoing monitoring
    • What characterizes a process as a model
    • What does inherent uncertainty imply about what we must assume?
    • Qualitative components vs. quantitative components
    •  How do machine learning models fit into the traditional MRM framework?
      • Commonality between data scientists and traditional statisticians
    • Accountability of the model owner and developer and proper documentation
  • Managing new processes with real time re-training and reporting requirements
    •  ML special features
    • Optimization for predictive performance
    • Highly automated/online learning
    • Complex representation
    • Stochastic training
    • Several hyperparameters
    • Open source and vendor ecosystem
    • Process automation applications
  • These requirements should boil into the following MRM themes:
    • Data bias and limitations
    • Explanability or interpretability
    • Replicability and stability
    • Ongoing monitoring 
    • Safety in autonomous mode
    • Fairness

Sean Naismith, GM & Head of Analytics Services
Enova Decisions
  • Non-linear, non-parametric model form: Important to capture the interaction effects & non-linearities inherent in fraud data
  • Automated model retraining: As fraud evolves, automated retraining of fraud ML models will pick up the predictive patterns left behind by fraudsters in the data
  • Velocity features: Leverage stream analytics to measure the frequency of a given variable or combination of variables within a specified time period as compared to a baseline period
  • Link variables: Leverage graph theory to connect quickly understand how attributes of applications & accounts are related and if there are risks
  • 3rd-Party Data: Go beyond internal data to get fraud signals from consortium providers

Who Should Attend

  • Financial Institutions
  • Fraud Analysts
  • Information Technology Executives
  • Fraud Monitoring Officers
  • Fraud Consultants
  • Privacy and Data Security Officers
  • In-house and Outside Counsel
  • Top Level Management

Online CLE Identity Fraud Prevention

Kevin King, Head of Marketing
ID Analytics, a Symantec Company

Clyde Langley, Vice President, Fraud Risk Management, Financial Crimes Risk Management
Charles Schwab & Co., Inc.
  • Consistent and accurate data feeds from multiple datapoints are the critical first step to implementing ML for fraud prevention and detection
  • ML requires sufficient activity and fraud instances to “learn” normal vs anomalous activity and detect fraud
  • Identify the right datapoints and ingest clean data for the model to learn normal activity
  • Fraud incident rates in the channel need to be sufficient to develop better predictive capabilities to detect fraud through ML
  • Understanding of ML fraud detection implementation decisions
  • ML implementation approach should be based on in-house capabilities (Wayne, this will likely tie-in directly based on your expertise
  • Consider criminals’ ability to identify and bypass your anti-fraud controls in implementing ML
  • Benefits (selling points) of ML in anti-fraud prevention and detection can reduce alerts and minimize operational costs while improving CX and mitigating fraud

Wayne Shoumaker, SVP, Quantitative Analytics Manager
Wells Fargo
  • Identity fraud spans a broad range of threats
    • Fraud models once considered outside of the scope of model risk management (MRM) are now considered in scope
    • Best practices need to align with FRB SR 11-7 and MRM
    • Regulatory guidance for a model’s conceptual soundness, outcomes analysis, and ongoing monitoring
    • What characterizes a process as a model
    • What does inherent uncertainty imply about what we must assume?
    • Qualitative components vs. quantitative components
    •  How do machine learning models fit into the traditional MRM framework?
      • Commonality between data scientists and traditional statisticians
    • Accountability of the model owner and developer and proper documentation
  • Managing new processes with real time re-training and reporting requirements
    •  ML special features
    • Optimization for predictive performance
    • Highly automated/online learning
    • Complex representation
    • Stochastic training
    • Several hyperparameters
    • Open source and vendor ecosystem
    • Process automation applications
  • These requirements should boil into the following MRM themes:
    • Data bias and limitations
    • Explanability or interpretability
    • Replicability and stability
    • Ongoing monitoring 
    • Safety in autonomous mode
    • Fairness

Sean Naismith, GM & Head of Analytics Services
Enova Decisions
  • Non-linear, non-parametric model form: Important to capture the interaction effects & non-linearities inherent in fraud data
  • Automated model retraining: As fraud evolves, automated retraining of fraud ML models will pick up the predictive patterns left behind by fraudsters in the data
  • Velocity features: Leverage stream analytics to measure the frequency of a given variable or combination of variables within a specified time period as compared to a baseline period
  • Link variables: Leverage graph theory to connect quickly understand how attributes of applications & accounts are related and if there are risks
  • 3rd-Party Data: Go beyond internal data to get fraud signals from consortium providers

Online CLE Identity Fraud Prevention

Online CLE Identity Fraud Prevention

Kevin KingHead of MarketingID Analytics, a Symantec Company

Kevin King has nearly a decade of experience in credit risk and fraud analytics, having served in a variety of roles at ID Analytics since joining the company in 2007 including analytics, business analysis, product strategy and professional services. In each of these roles, King has focused on applying ground breaking analytic tools to the unique financial services challenges, and has directly supported projects at top 10 banks, several “Big 4” wireless carriers, the two leading U.S. satellite television companies, and multiple leading cable and internet providers. A driving force behind ID Analytics’ thought leadership program, he has authored several thought leadership whitepapers on a range of topics spanning fraud, credit and identity risk. Kevin has a B.A. in Marketing from the University of Colorado at Boulder.

Online CLE Identity Fraud Prevention

Clyde LangleyVice President, Fraud Risk Management, Financial Crimes Risk ManagementCharles Schwab & Co., Inc.

Clyde Langley is the Vice President, Head of Fraud Risk Management, for Charles Schwab & Co., Inc. The Fraud Risk Management group is responsible for firm-wide Fraud Investigations, including Cybercrime Detection and Investigations (online fraud), Anti-Fraud Digital Strategy, Fraud Strategies & Surveillance, Fraud & Security Awareness, Visa & Check Fraud and Fraud Investigations teams. He also oversees the Identity Theft Red Flags Program.

Before joining Schwab in 2014, Clyde spent 22 years as an FBI Special Agent in Executive Management to Investigator roles. In the FBI, Clyde spent over 15 years supervising and investigating financial and cybercrimes, including 8 years managing financial crimes investigations and four years overseas in Kazakhstan and Belgium. After the 9/11 attacks, Clyde was part of the team that established the FBI’s terrorist financing section to enhance terrorism investigations from the financial perspective. Prior to the FBI, Clyde spent five years as an auditor/accounting manager in public accounting and private industry.

Online CLE Identity Fraud Prevention

Wayne ShoumakerSVP, Quantitative Analytics ManagerWells Fargo

Online CLE Identity Fraud Prevention

Sean NaismithGM & Head of Analytics ServicesEnova Decisions

Sean joined Enova as Head of Analytics Services for Enova Decisions in 2016. Prior to working at Enova, Sean served as Senior Director of Business Analytics for Leapfrog, where he led the development of the company’s predictive analytics capabilities. Before Leapfrog, Sean served as Director of Strategic Intelligence for TrendPointers, LLC, and Associate Portfolio Manager at Sarasota Capital Strategies. Sean is a CFP® certificant and holds the CMT designation. He received his B.S. in finance from the University of Illinois at Chicago and Financial Planning Certificate from Northwestern University.

Online CLE Identity Fraud Prevention

Course Level:
   Intermediate

Advance Preparation:
   Print and review course materials

Method Of Presentation:
   Live Webcast

Prerequisite:
   General knowledge of financial fraud

Course Code:
   148511

NY Category of CLE Credit:
   Areas of Professional Practice

Total Credits:
    2.0 CLE

No Access

You are not logged in. Please or register to the event to gain access to the materials and login instructions.

About the Knowledge Group

The Knowledge Group

The Knowledge Group has been a leading global provider of Continuing Education (CLE, CPE) for over 13 Years. We produce over 450 LIVE webcasts annually and have a catalog of over 4,000 on-demand courses.

About the Knowledge Group

The Knowledge Group

The Knowledge Group has been a leading global provider of Continuing Education (CLE, CPE) for over 13 Years. We produce over 450 LIVE webcasts annually and have a catalog of over 4,000 on-demand courses.

ID Analytics is a leader in consumer risk management with patented analytics, proven expertise and near real-time insight into consumer behavior. By combining proprietary data from the ID Network®—one of the nation’s largest networks of cross-industry consumer behavioral data—with advanced science, ID Analytics provides in-depth visibility into identity and credit risk. Every day, many of the largest U.S. companies rely on ID Analytics to make risk-based decisions that help enhance revenue, reduce fraud, drive cost savings and protect consumers. ID Analytics is a Symantec company. Please visit us at www.idanalytics.com.

Website: https://www.idanalytics.com/

Wells Fargo has become one of the nation's largest financial institutions, serving one in three households in the United States. We provide banking, investments, mortgage, and consumer and commercial finance through 7,800 locations, more than 13,000 ATMs, the internet (wellsfargo.com), and mobile banking, and we have offices in 37 countries and territories to support customers who conduct business in the global economy. Wells Fargo & Company was ranked No. 26 on Fortune's 2018 rankings of America's largest corporations.

Website: https://www.wellsfargo.com/

Enova Decisions is an analytics and decision management technology company that was formed in 2016 to enable businesses to automate and optimize operational decisions through data, AI, and the cloud — in real-time and at scale. Enova Decisions is part of Enova International, Inc. (NYSE: ENVA), a Chicago-based multinational financial services provider that has applied predictive and prescriptive analytics to multiple areas, including fraud detection, credit risk management and customer retention, to extend over $20 billion in credit online to over 5 million customers. Enova Decisions leverages the same analytics expertise and decisioning technology that has made Enova International successful to help businesses in multiple industries, including financial services, healthcare and telecommunications, achieve similar outcomes.

Website: https://www.enovadecisions.com/

Kevin King has nearly a decade of experience in credit risk and fraud analytics, having served in a variety of roles at ID Analytics since joining the company in 2007 including analytics, business analysis, product strategy and professional services. In each of these roles, King has focused on applying ground breaking analytic tools to the unique financial services challenges, and has directly supported projects at top 10 banks, several “Big 4” wireless carriers, the two leading U.S. satellite television companies, and multiple leading cable and internet providers. A driving force behind ID Analytics’ thought leadership program, he has authored several thought leadership whitepapers on a range of topics spanning fraud, credit and identity risk. Kevin has a B.A. in Marketing from the University of Colorado at Boulder.

Clyde Langley is the Vice President, Head of Fraud Risk Management, for Charles Schwab & Co., Inc. The Fraud Risk Management group is responsible for firm-wide Fraud Investigations, including Cybercrime Detection and Investigations (online fraud), Anti-Fraud Digital Strategy, Fraud Strategies & Surveillance, Fraud & Security Awareness, Visa & Check Fraud and Fraud Investigations teams. He also oversees the Identity Theft Red Flags Program.

Before joining Schwab in 2014, Clyde spent 22 years as an FBI Special Agent in Executive Management to Investigator roles. In the FBI, Clyde spent over 15 years supervising and investigating financial and cybercrimes, including 8 years managing financial crimes investigations and four years overseas in Kazakhstan and Belgium. After the 9/11 attacks, Clyde was part of the team that established the FBI’s terrorist financing section to enhance terrorism investigations from the financial perspective. Prior to the FBI, Clyde spent five years as an auditor/accounting manager in public accounting and private industry.

Sean joined Enova as Head of Analytics Services for Enova Decisions in 2016. Prior to working at Enova, Sean served as Senior Director of Business Analytics for Leapfrog, where he led the development of the company’s predictive analytics capabilities. Before Leapfrog, Sean served as Director of Strategic Intelligence for TrendPointers, LLC, and Associate Portfolio Manager at Sarasota Capital Strategies. Sean is a CFP® certificant and holds the CMT designation. He received his B.S. in finance from the University of Illinois at Chicago and Financial Planning Certificate from Northwestern University.

Ultimate Value Annual Program

Bring a colleague for only $149, a savings of $50 per additional attendee.

  • Unlimited Access to Live & Recorded Webcasts
  • Instant Access to Course Materials
  • And More!

$199