Advanced Analytics Lab
SAS Fraud Framework and Social Network Analysis For Insurance
Copyright © 2009, SAS Institute Inc. All rights reserved.
Advanced Analytics Lab
Starting with the SAS Fraud Framework Increasing Fraud - The Business Problem Fraudsters • • • • •
Far more sophisticated – organized crime, patient, sharing of rules Engaging insiders to understand detection environment High velocity of attacks – disappear after 2-3 transactions Hitting multiple channels and industries at the same time Continuously evolving fraud strategies
Current Fraud Systems • • • • • •
Systems are silo’d by line of business - no sharing of data Current systems act on transaction or customer Rules and predictive models alone have limitations Rely on 3rd party systems No real proactive steps taken to combat Fraud (or Churn) Evidence insufficient to act upon 2
Copyright © 2009, SAS Institute Inc. All rights reserved.
Advanced Analytics Lab
SAS Fraud Framework Innovation in detection driven by industry Robust and flexible framework capabilities • Support for real-time, intra-day, batch execution
• Ability to use existing data infrastructure • Ability to use existing fraud alert output from any LOB / 3rd party • Business intelligence for all levels of users
Support for business functions • Provide strategic insight into threats, trends, risks • Enterprise view of fraudulent behavior • Rapidly test , simulate, and deploy models/rules without dependence on IT • Ability to provide single view for investigators
Phased approach to support tactical & strategic initiatives 3 Copyright © 2009, SAS Institute Inc. All rights reserved.
Advanced Analytics Lab
Customer Successes Large Auto Insurer (PIP) 100,000 claims supplied from high risk geo in historical data set Over $70M in total payments on Investigation Worthy alerts identified in the validation phase
Regional Auto Insurer Used 27 months of historical data from one PIP/casualty jurisdiction
Over $11M in actionable dollars-at-risk identified in validation phase
4 Copyright © 2009, SAS Institute Inc. All rights reserved.
Advanced Analytics Lab
SAS Fraud Framework Using a Hybrid Approach for Fraud Detection
5 Copyright © 2009, SAS Institute Inc. All rights reserved.
Advanced Analytics Lab
Analytic Engine Analytic Approach: Unsupervised Methods Use when no target exists Examine current behavior to identify outliers and abnormal transactions that are somewhat different from ordinary transactions Include univariate and multivariate outlier detection techniques, such as peer group comparison, clustering, trend analysis, and so on 6 Copyright © 2009, SAS Institute Inc. All rights reserved.
Advanced Analytics Lab
Analytic Engine Analytic Approach: Supervised Methods Use when a known target (fraud) is available Use historical behavioral information of known fraud to identify suspicious behaviors similar to previous fraud patterns
Fraud Scores
Predicted Fraud Scores Incomes
# of previous investigations
Include parametric and nonparametric predictive models, such as generalized linear model, tree, neural networks, and so on 7 Copyright © 2009, SAS Institute Inc. All rights reserved.
Advanced Analytics Lab
Analytic Engine Predictive Analytics
8 Copyright © 2009, SAS Institute Inc. All rights reserved.
Advanced Analytics Lab
Analytic Engine Analytic Approach: Text Mining
Text Mining (e.g., call center logs or investigator notes) 9 Copyright © 2009, SAS Institute Inc. All rights reserved.
Advanced Analytics Lab
Analytic Engine SAS Social Network Analysis Network scoring • Rule and analytic-based
Analytic measures of association help users know where to look in network • Net-CHAID for local area of interest (node) in the network • Density, Beta-Index (network) • Risk ranking with hypergeometric distribution, degree, closeness, betweenness, eigenvector, clustering coefficients (node)
Modularity (sub-network) 10 Copyright © 2009, SAS Institute Inc. All rights reserved.
Advanced Analytics Lab
SAS® Fraud Framework Process Flow Operational Data Sources
Exploratory Data Analysis & Transformation
Alert Generation Process Fraud Data Staging
Customers
Business Rules
Alert Administration
SAS® Social Network Analysis
Analytics
Network Rules
Anomaly Detection
Network Analytics
Predictive Modeling Accounts
Transactions
Intelligent Fraud Repository
Learn and Improve Cycle
Alert Management & Reporting Case Management
11 Copyright © 2009, SAS Institute Inc. All rights reserved.
Advanced Analytics Lab
Why Social Network Analysis? Provides the ability to apply Rules, Predictive Models, and Anomaly Detection on linked data More fraud/actionable cases identified • Including both previously undetected networks and extensions to already identified cases
Reduction in false positive rates • SNA reduces false positives by up to 10+ times over traditional rulesbased approaches
Improved analyst / investigation efficiency • Each referral takes 1/2 – 1/3 the time to investigate using SAS’ fraud network visualization on aggregated data
Significant increase in ROI per analyst / investigator 12 Copyright © 2009, SAS Institute Inc. All rights reserved.
Advanced Analytics Lab
Demo Screenshots
13 Copyright © 2009, SAS Institute Inc. All rights reserved.
Advanced Analytics Lab
Alerts Page
14 Copyright © 2009, SAS Institute Inc. All rights reserved.
Advanced Analytics Lab
Provider Details
15 Copyright © 2009, SAS Institute Inc. All rights reserved.
Advanced Analytics Lab
Details Page
16 Copyright © 2009, SAS Institute Inc. All rights reserved.
Advanced Analytics Lab
Network
17 Copyright © 2009, SAS Institute Inc. All rights reserved.
Advanced Analytics Lab
Demo Movie Social Network Analysis – In Action http://tinyurl.com/SFFDemoMovie
18 Copyright © 2009, SAS Institute Inc. All rights reserved.
Advanced Analytics Lab
19 Copyright © 2008, 2009, SAS Institute Inc. All rights reserved.