Full Agenda – October 19-22, 2009
sessions are for all levels.
sessions are expert/practitioner level.

Monday October 19, 2009

Full-day Workshop
Room: Poplar
Putting Predictive Analytics to Work

  • Workshop starts at 9:00am
  • Morning Coffee Break at 10:30am - 11:00am
  • Lunch provided at 12:30 - 1:15pm
  • Afternoon Coffee Break at 2:30pm - 3:00pm
  • End of the Workshop: 4:30pm

Speaker: James Taylor, CEO, Decision Management Solutions


Full-day Workshop
Room: Walnut B
Hands-On Predictive Analytics

  • Workshop starts at 9:00am
  • Morning Coffee Break at 10:30am - 11:00am
  • Lunch provided at 12:30 - 1:15pm
  • Afternoon Coffee Break at 2:30pm - 3:00pm
  • End of the Workshop: 4:30pm

Speaker: Dean Abbott, President, Abbott Analytics




Tuesday October 20, 2009

8:00am-9:00am
Registration & Continental Breakfast


9:00am-9:50am
Keynote
Room: Magnolia
Five Ways to Lower Costs with Predictive Analytics

Question: How does predictive analytics actively deliver increased returns? Answer: By driving operational decisions with predictive scores - one score assigned to each customer. In this way, an enterprise optimizes on what customers WILL do.

But, in tough times, our attention turns away from increasing returns, and towards decreasing costs. On top of boosting us up the hill, can predictive analytics pull us out of a hole? Heck, yes. Marketing more optimally means you can market less. Filtering high risk prospects means you will spend less. And, by retaining customers more efficiently, well, a customer saved is a customer earned - and one you need not acquire.

In this keynote, Eric will demonstrate five ways predictive analytics can lower costs without decreasing business, thus transforming your enterprise into a Lean, Mean Analytical Machine. You'll want to run back home and break the news: We can't afford not to do this.

Speaker: Eric Siegel, Ph.D., Conference Chair


9:50am-10:10am
Platinum Sponsor Presentation
Room: Magnolia
The Perfect Storm: The Rise of Predictive Analytics

In today's world, a unique combination of trends and factors is driving the uptake of predictive analytics in organizations across all sectors. The explosion of data volumes and the availability of advanced analytical technology coincide with an unprecedented focus on generating return on investments (ROI) in business systems and processes.

This presentation will address issues such as:

  • How mathematics is becoming a trusted and popular science
  • How the amount of data available is exponential exploding
  • How the new sources of data are contributing to that explosion
  • How C-level executives are pushing predictive analytics to the forefront
  • How predictive analytics techniques are redefining the role of IT
  • How organizations justify their investments in predictive analytics

Moderator: Eric Siegel, Program Chair, Predictive Analytics World

Speaker: Erick Brethenoux, VP Corporate Development, SPSS Inc.


10:10am-10:30am
Morning Coffee Break
Upper Foyer / Exhibit Hall

Exhibit Hall Open from 10:10am to 7:30pm


Delegates may choose to attend either session at this time.

10:30am-11:20am
Track 1: Thought Leader
Room: Magnolia
Case Study: Infinity Insurance & PREMIER Bankcard
Putting Predictive Analytics to Work

Seeking out increasingly small margins has brought analytics into vogue. Organizations realize that they can gain analytic insights about customers, products, channels, partners and much more. But some companies are already finding that analytics is only a part of the process - the intelligent application of the findings of these new insights can only pay off if the decisions that are made are correct.

Translating analytics into better operational outcomes requires a new conceptual framework - decision management. By applying this framework and becoming more decision-centric, by using business rules to control those decisions and by leveraging predictive analytics, companies are truly putting predictive analytics to work.

You will learn:

  • How better operational decisions deliver ROI
  • What the challenges are in applying predictive analytics to these decisions
  • What Decision Management is and how it addresses these challenges
  • How companies have benefited from this approach

Moderator: Eric Siegel, Program Chair, Predictive Analytics World

Speaker: James Taylor, CEO, Decision Management Solutions

10:30am-11:20am
Track 2: Non-profit
Room: Walnut A&B
Case Study: National Rifle Association
How to Improve Customer Acquisition Models with Ensembles

Model ensembles have been used increasingly by data mining practitioners to increase model accuracy and routinely appear as top performers in predictive modeling competitions. However, they also provide significant robustness advantages over single models, particularly when data is noisy and has significant levels of uncertainty.

Insightful Miner was used to create an ensemble of models to improve the ROI of a direct marketing campaign to members of the National Rifle Association. This session will highlight a systematic methodology to use in building bootstrapped ensembles of logistic regression models, similar to the Bagging approach first advocated by Leo Brieman. The model ensembles are more reliable and out perform single models by exploring the results of applying both modeling techniques to the NRA membership file.

Moderator: Karl Rexer, Rexer Analytics

Speaker: Dean Abbott, President, Abbott Analytics


11:20am-12:30pm
Multiple Case Studies: Anheuser-Busch, Disney, HSBC, Pfizer, and others
Room: Magnolia
The High ROI of Data Mining for Innovative Organizations

Data mining and advanced analytics can enhance your bottom line in three basic ways, by 1) streamlining a process, 2) eliminating the bad, or 3) highlighting the good. In rare situations, a fourth way - creating something new - is possible. But modern organizations are so effective at their core tasks that data mining usually results in an iterative, rather than transformative, improvement. Still, the impact can be dramatic.

Dr. Elder will share the story (problem, solution, and effect) of nine projects conducted over the last decade for some of America's most innovative agencies and corporations:

    Streamline:
  • Cross-selling for HSBC
  • Image recognition for Anheuser-Busch
  • Biometric identification for Lumidigm (for Disney)
  • Optimal decisioning for a leading high-tech retailer
  • Quick decisions for the Social Security Administration
    Eliminate Bad:
  • Tax fraud detection for the IRS
  • Warranty Fraud detection for a leading high-tech retailer
    Highlight Good:
  • Sector trading for WestWind Foundation
  • Drug efficacy discovery for Pharmacia & UpJohn (now Pfizer)

Moderator: Eric Siegel, Program Chair, Predictive Analytics World

Speaker: John Elder, Elder Research, Inc.


12:30pm-1:45pm
Birds of a Feather Lunch
Room: Pan Am Foyer

Find your clan of like-minded colleagues with whom to dine and discuss, comparing your organization's stories and challenges.

Discussion topics:

  • Response modeling
  • Churn modeling
  • Product recommendations

SAS Lunch Topics:

  • Response modeling
  • Churn modeling
  • Interactive data visualization best practices

1:45pm-2:35pm
Keynote
Room: Magnolia
Predictive Analytics over On-line and Social Network Data

The rise of the interactive media represented by web, social media, search and behavioral targeting have created new challenges and opportunities for predictive analytics. While these new media offer a better chance for approaching the holy grail in marketing and advertising - understanding the customer's intent and utilizing this intent to produce relevant offers and advertising - the rich structure of this data, from social graph data to time-series from interactions to reputation and other behavioral traits, from video streams to unstructured text, expand the complexity of prediction in dimensions where we have little experience and a poor understanding of the terrain. We present examples of such applications as well as challenges, and relate some case-studies to illustrate the power of understanding and harnessing this data. However, the context will also be used to illustrate the stronger need to build up new sciences to help us better understand these new powerful dimensions and new areas of their application. Predictive analytics plays a key role in addressing many of these challenges.

Moderator: Eric Siegel, Program Chair, Predictive Analytics World

Speaker: Usama Fayyad, Ph.D., Open Insights - Former Chief Data Officer, Yahoo!


2:35pm-2:45pm
Leveraging Speech Analytics to Gain a Competitive Edge
Room: Magnolia

In this overview, learn how NICE has become the leading provider of Insight from Interactions solutions, powered by advanced analytics of unstructured multimedia content - from telephony, web, radio and video communications. NICE's solutions address the needs of the enterprise market enabling organizations to operate in an insightful and proactive manner. Every conversation between your customers and staff speaks volumes about your company's long-term success. With Interaction Business Analytics from NICE Systems, you can tune into the valuable data housed in every interaction to establish sound business strategies for delivering exemplary service and shaping future strategy.

Moderator: Eric Siegel, Program Chair, Predictive Analytics World

Speaker: Russell Mandelik, NICE Systems


2:45pm-2:55pm
Session Break
Upper Foyer / Exhibit Hall


Delegates may choose to attend either session at this time.

2:55pm-3:45pm
Track 1: Incremental Modeling (Uplift modeling)
Room: Magnolia
Case Study: Target
Challenges of Incremental Sales Modeling in Direct Marketing

While Target has enjoyed profitable returns from their ongoing direct mail program, the Target analytics group was given the challenge to drive more guest (customer) incremental sales and profit in the future. The team determined that in order to do this, they needed to look beyond the common levers like guest segmentation, product category guest conversion modeling, and mail piece customization to tackle guest-level incremental sales modeling.

In this presentation, we will share our proposed methodologies, as well as results, based on using predictive models to identify guests who are likely to spend incrementally upon receiving a direct mail contact. Unfortunately, despite a year's worth of effort and multiple engagements with statistical consulting firms, the team has been unable to produce consistent and dependable result to use in guest campaign selection. Given this, the Target team will share intermediate results and cover hypotheses as to why the solution has been so elusive to date.

Moderator: Eric Siegel, Program Chair, Predictive Analytics World

Speaker: Andrew Pole, Senior Manager, Media and Database Marketing, Target

2:55pm-3:45pm
Track 2: Telecommunications
Room: Walnut A&B
Case Study: Sunrise Communications (Switzerland)
Cost Reduction in Bill-Insert Campaigns With Predictive Analytics

Sunrise Communications is the 2nd largest telecom provider in Switzerland. Its business intelligence team has successfully deployed churn modeling for several years. In this session, we describe a new data mining application within Sunrise's CRM department.

Every month, CRM creates several print campaigns which are sent to the customers together with their invoices. To reduce costs, however, not every customer receives a monthly invoice. Whether the invoice is sent depends on a complex calculation based on the month's revenues, any outstanding amounts, previous bills, etc.

The challenge arises from the fact that CRM has to calculate the volumes to be sent to the printer several weeks before the month's revenues is known. Predicting the required print volumes has resulted in significant cost savings compared to the approach taken previously.

Moderator: Eric A. King, The Modeling Agency

Speaker: Stamatis Stefanakos, Senior Consultant, D1 Solutions AG


Delegates may choose to attend either session at this time.

3:45pm-4:30pm
Track 1: Incremental Modeling (Uplift modeling)
Room: Magnolia
Case Study: US Bank
Raising the Bar in Cross-Sell Marketing With Uplift Modeling

Learn how US Bank is applying next generation Uplift analytic modeling to boost the ROI of their cross-sell marketing while simultaneously slashing program costs. As you will hear, Uplift modeling significantly exceeded the results achievable via traditional analytic approaches. Some of the many results observed include:

  • Increase incremental cross-sell revenue by greater than 300%
  • Reduce mailing costs by up to 40%
  • Isolate and eliminate the negative effects of marketing
  • Achieve a 5-fold increase in campaign ROI when compared with existing programs

In this session you will also receive practical tips from experts on how you can leverage this technique within your organization in order to optimize marketing performance by achieving greater business results from increasingly limited resources.

Moderator: Eric Siegel, Program Chair, Predictive Analytics World

Speaker: Michael Grundhoefer, Marketing Analytics, Market Info. & Research, US Bank

3:45pm-4:30pm
Track 2: Telecommunications
Room: Walnut A&B
Case Study: Optus (Australian telecom)
Know Your Customers by Knowing Who They Know, and Who They Don't

In the highly saturated and ever competitive mobile (cell-phone) industry, telecommunication companies continually seek new ways to analyse and understand customer behaviour. Whilst the priority of data mining analysis in the telecommunications industry is often to retain customers, being able to successfully target customers for appropriate upgrades (for example, new handsets or higher priced plans) and additional value added services will greatly enhance customer value and revenue profit margins. One recent development in Optus SingTel's strategy for better customer insights is social network analysis

Tim Manns will present a case study of an In-Database Social Networking Analysis solution developed and applied to the Optus SingTel consumer mobile customer base (approx 5 million mobile customers). This deployed data mining solution runs monthly, analysing every communication event made and received by mobile customers. The presentation will illustrate some of the practical data mining challenges of processing and transforming several billion rows of data warehouse transactional level data (call detail records) into summarised customer data within a practical timeframe. Business benefits and predictive modelling outcomes will also be presented.

Moderator: Eric A. King, The Modeling Agency

Speaker: Tim Manns, Data Miner, Optus (part of the SingTel group)


4:30pm-4:55pm
Afternoon Coffee Break
Upper Foyer / Exhibit Hall


Delegates may choose to attend either session at this time.

4:55pm-5:40pm
Track 1: Financial Services
Room: Magnolia
Case Study: PREMIER Bankcard
The Development of a "Good Customer Score" for Use in Customer Acquisition, Rewards, Retention and Recovery

"Good Customer Score (GCS)" is a measurement of current customer value. The methodology used for development of GCS incorporates internal operational customer data focused on key customer performance measures applied through mathematical weighting to generate a ratio representative of a "Good Customer".

The GCS accuracy is supported by its statistical correlation to Behavior Score (3rd party score), as well as other scores, when predicting those customers who will perform in the top 25% of the portfolio ranked by GCS. The strength of like scores is measured using Chi-Square correlation results in addition to Decision Tree and Logistic Regression prediction models.

Moderator: Eric Siegel, Program Chair, Predictive Analytics World

Speaker: Rex Pruitt, Senior Level Business Analyst, PREMIER Bankcard, LLC

4:55pm-5:40pm
Track 2: Telecommunications
Room: Walnut A&B
KDDcup 2009 Competition Results: Orange Labs (France Telecom)
Churn, Baby, Churn: Fast Scoring on Large Telecom Dataset

Churn prediction and management is critical for companies in the fast and competitive telecommunication market. In this session, we introduce the dataflow computational model in the context of data and computationally intensive high performance parallel data mining. We present a highly scalable and robust model capable of scoring "propensity-to-churn" at the rate of 50,000 customers in a 1.6GB test set (Orange Labs France Telecom, KDD Cup) in 3 minutes on commodity 16-core CPUs. This is an effective scoring runtime of 3.6 milliseconds per customer, orders of magnitude faster than some systems. As a competitor in this year's KDDcup data mining competition, this speed enables more iterations towards improved performance; while the research focus was speed, resulting predictive accuracy ranked higher than 70% of competitors.

Moderator: Eric A. King, The Modeling Agency

Speaker: Srivatsava Daruru, Research Assistant, The University of Texas at Austin


Delegates may choose to attend either session at this time.

5:40pm-6:30pm
Track 1: Financial Services
Room: Magnolia
Case Study: Citizens Bank
Building In-Database Predictive Scoring Model: Check Fraud Detection Case Study

It is estimated that the nation's banks experience over $10 billion per year in attempted check fraud. The daily challenge for a large bank is to identify a few thousand risky checks deposited out of hundreds of thousands of normal ones. To address this challenge, we build a predictive model that gives each check a risk score. All of the processes including data preparation, feature variable calculation, model training, model testing and final model deployment are executed within a single database environment. The in-database solution provides substantially increased security, productivity, manageability and scalability.

Moderator: Eric Siegel, Program Chair, Predictive Analytics World

Speaker: Jay Zhou, President, Business Data Miners, LLC

5:40pm-6:30pm
Track 2: Text Analytics
Room: Walnut A&B
Predictive Text Analytics

The session will introduce predictive analytics applied to textual material and present several mini-application case studies. Predictive text analytics consists of information retrieval, information extraction, clustering, classification, and other techniques applied to make sense of 'unstructured' sources. Solutions apply statistical, linguistic, and machine learning algorithms in conjunction with BI, data mining, and visualization tools.

The technology has found broad applications in business, government, and research, in domains that range from intelligence and the life sciences to social-media analysis. The case studies will highlight challenges, techniques, and benefits in the application of predictive text analytics in a sampling of the many business domains where solutions are applied.

Moderator: Eric A. King, The Modeling Agency

Speaker: Seth Grimes, Principal Consultant, Alta Plana Corporation


6:30pm-7:30pm
Reception
Upper Foyer / Exhibit Hall

Sponsored by  


7:30pm-10:00pm
useR Meeting
Room: Magnolia
- Sponsored by  

Please join the group at www.meetup.com/R-users-DC/

R is an open source programming language for statistical computing, data analysis, and graphical visualization. R has an estimated one million users worldwide, and its user base is growing. While most commonly used within academia, in fields such as computational biology and applied statistics, it is gaining currency in commercial areas such as quantitative finance and business intelligence.

Among R's strengths as a language are its powerful built-in tools for inferential statistics, its compact modeling syntax, its data visualization capabilities, and its ease of connectivity with persistent data stores (from databases to flatfiles).

In addition, R is open source nature and extensible via add-on "packages" allowing it to keep up with the leading edge in academic research.

For all its strengths, though, R has an admittedly steep learning curve; the first steps towards learning and using R can be challenging.

This DC R Users Group is dedicated to bringing together area practitioners of R to exchange knowledge, inspire new users, and spur the adoption of R for innovative research and commercial applications.




Wednesday October 21, 2009

8:00am-9:00am
Registration & Continental Breakfast


9:00am-9:50am
Keynote
Room: Magnolia
Opportunities and Pitfalls:
What the World Does and Doesn't Want from Predictive Analytics

Mathematicians and statisticians are churning through mountains of data in their efforts to model and predict human behavior. The goal is to optimize every function possible, from sales and marketing to the enterprise itself. These Numerati are guided by the two dominant models of the late 20th century, the modeling of financial markets and of industrial systems. How do humans fit into these systems? And what will their response be when the analytic systems appear to misunderstand them or invade their privacy?

Stephen Baker joins PAW to directly address the Numerati. In his keynote presentation, Mr. Baker will guide us toward the untapped goldmines where predictive analytics will be embraced and thrive, and teach us to anticipate and maneuver around two central pitfalls: Consumer misperception of us, and our inadvertent mistreatment of them.

Moderator: Eric Siegel, Program Chair, Predictive Analytics World

Speaker: Stephen Baker, BusinessWeek - author, The Numerati


9:50am-10:10am
Platinum Sponsor Presentation
Room: Magnolia
Strength in Numbers: ACE!

As more organizations are beginning their analytical journey or reinvigorating their existing efforts, Analytic Centers of Excellence (ACEs) are helping them along the way. The interest in ACEs is growing across industries as organizations seek better ways to tap into their analytic infrastructure-most importantly, scarce high-end analytic expertise to improve results. We will highlight valuable best practices for achieving greater analytic bandwidth realizing more and better evidence-based decisions.

Moderator: Eric Siegel, Program Chair, Predictive Analytics World

Speaker: Anne Milley, Senior Director of Tech. Product Marketing, SAS


10:10am-10:30am
Morning Coffee Break
Upper Foyer / Exhibit Hall

Exhibit Hall Open from 10:10am to 6:00pm


Delegates may choose to attend either session at this time.

10:30am-11:20am
Track 1: Verticals
Room: Magnolia
Case Study: Reed Elsevier
Where Do We Go from Here - So the First Model Worked. What About the Next 6?

At the inaugural PAW in February 2009 John presented a first case study in subscriber retention modelling. The emphasis of this project was to improve retention rates and profitability for one of Reed Business Information's magazines, Caterer & Hotelkeeper, and to do this in a way which enabled Ed Garcia of RBI to prove the business value of the resultant model to senior management.

Since then we've moved on from the initial success and modelled a number of other RBI publications including New Scientist and Flight International. Our new case study presents the results from the next level of modelling.

Moderator: Karl Rexer, Rexer Analytics

Speaker: John McConnell, Director, Analytical People

10:30am-11:20am
Track 2: Forecasting
Room: Walnut A&B
Case Study: The Financial Times, The New York Times, Sprint-Nextel
Predicting Future Subscriber Levels

Subscription-based businesses of all types--software as a service firms, wireless carriers, satellite and cable TV firms--all have a need to predict future subscriber levels. This session presents a customer-centric approach to prediction using survival analysis. A subscriber population is an ever-changing mix of customer segments, each with its own hazard probability that is a function of tenure and additional covariates such as market, product type, and credit class. A forecast based on these hazard probabilities automatically reacts to changes in the customer mix, allowing one to simulate alternate scenarios based on different assumptions about future customer acquisitions.

Moderator: Eric Siegel, Program Chair, Predictive Analytics World

Speaker: Michael Berry, Founder and Principal, Data Miners, Inc.


Delegates may choose to attend either session at this time.

11:20am-12:10pm
Track 1: Verticals
Room: Magnolia
Case Study: Amway
Establishing a Performance-Based Culture with Predictive Analytics

Too often organizations manage their business based on what they know from the past. In Amway's case, they use Predictive Analytics as a necessary and strategic business weapon to provide the vision to proactively make decisions that improve the future. Amway will discuss how Predictive Analytics has helped transform their organization into a performance-based culture. In support of this worldwide initiative, Amway developed key metrics and processes across numerous program, such as measuring the lifetime value of its Distributors.

By implementing Predictive Analytics, Amway was able to successfully execute this critical metric quickly by building, validating and deploying models more efficiently that benefited its marketing campaigns, segmentation, strategic planning and, most importantly, improved Distributor retention rates. Amway's management team now has clear insight into the results of this analysis on an executive dashboard so they can manage and measure investment decisions and track program success. Amway will explain the process, techniques and lessons learned from this initiative, including an overview of the metrics, roll-out stages, aligning with IT, communicating with management and successes.

Moderator: Karl Rexer, Rexer Analytics

Speaker: Mike Kinlaw, Manager/Lead Statistician, CMI Customer Analytics, Amway

11:20am-12:10pm
Track 2: Forecasting
Room: Walnut A&B
Case Study: Coke
A Predictive Approach to Marketing Mix Modeling

In this session we will review approaches to Marketing Mix Modeling with forecasting as a primary focus. We will review trade-offs in data granularity to achieve the accuracy desired to estimate the impact of various promotional activities on sales and then forecast the impact of future sets of promotional tactics with and eye towards different departments within the organization that are affected. We will address certain limitations such as reading the value of longer-term equity building programs and regional or targeted media impact. And finally, we will review the System Effort required to get worthwhile results in terms of action-ability.

Moderator: Eric Siegel, Program Chair, Predictive Analytics World

Speaker: Ram Krishnamurthy, Group Director, Marketing Strategy & Insights, Coca-Cola

Speaker: Anish Nanavaty, CEO, Research & Analytics, WNS Global Services, Inc


12:10pm-1:50pm
Birds of a Feather Lunch
Room: Pan Am Foyer

Find your clan of like-minded colleagues with whom to dine and discuss, comparing your organization's stories and challenges.

Discussion topics:

  • Project management and organizational process
  • Data preparation
  • Core predictive modeling methods

SAS Lunch Topics:

  • Project management and organizational process
  • Data access and preparation
  • Core predictive modeling methods

1:50pm-2:40pm
Expert Panel: Predictive Analytics and Consumer Privacy
Room: Magnolia

When analytics delivers value to an enterprise, it often means benefits for the end consumer as well. More precise targeting means less junk in the mail (not to mention fewer trees cut down). More effective fraud detection allows for more competitive pricing. Spot-on personal movie, music and book recommendations certainly can't hurt. What could possibly go wrong?

Pure and simple, personalization means personal data must be stored. Moreover, with predictive models in place, this personal data becomes alive as it is specifically acted upon. Today's consumer displays a growing concern with her "cloud identity". What's known - and what's thereby inferred - about a consumer seems to, at times, cut to the core of her sense of identity.

Our panel of experts digs in to answer the central questions. What happens when the shoppers and workers of the world perceive predictive analytics as irresponsibly applied - or, worse yet, what if they're right? What are the minimally required conditions to ensure the consumer does not feel manipulated?

Moderator: Eric Siegel, Program Chair, Predictive Analytics World

Speaker: Stephen Baker, BusinessWeek - author, The Numerati

Speaker: Jules Polonetsky, Co-Chair and Director, Future of Privacy Forum

Speaker: Mikael Hagström, Executive Vice President, EMEA and Asia Pacific, SAS


2:40pm-2:45pm
Session Break
Upper Foyer / Exhibit Hall


Delegates may choose to attend either session at this time.

2:45pm-3:35pm
Track 1: Health Care
Room: Magnolia
Case Study: Lifeline Screening
Segmented Modeling Applications in Health Care Industry

A "segmented" modeling methodology was developed in order to identify most likely responders for Lifeline Screening's Direct Mail program. The models are developed based on segments of the population that respond differently depending on the frequency and cadence of the communications. "Segment-specific" models were developed over different consumer segments . This approach produced significant lift over the traditional response modeling approach that is based on a single model, as established with a head-to-head evaluation.

Moderator: Karl Rexer, Rexer Analytics

Speaker: Ozgur Dogan, Vice President, Merkle

2:45pm-3:35pm
Track 2: Fraud Detection
Room: Walnut A&B
Keep Winning the Eternal Fraud Battles

Fraud is pervasive and extraordinarily costly, and the effort required to prevent it is non-trivial. But, enormous ROI is possible when predictive analytics insights are harnessed to detect ever-changing anomalous behavior. We will describe case studies to highlight lessons learned about what leads to a successful fraud detection project. We'll also address the cultural and business hazards of attacking versus ignoring fraud, summarize collateral benefits of analyzing customer behavior, and describe new forms of fraud only discoverable via analyzing social networks and links between accounts.

Moderator: Eric Siegel, Program Chair, Predictive Analytics World

Speaker: Antonia de Medinaceli, Senior Business Analyst, Elder Research, Inc.


3:35pm-3:55pm
Afternoon Coffee Break
Upper Foyer / Exhibit Hall


Delegates may choose to attend either session at this time.

3:55pm-4:45pm
Track 1: Insurance
Room: Magnolia
Case Study: Zurich
Top 10 Ways to be Successful in Implementing Predictive Modeling in Insurance Commercial Markets

Zurich recently chalked up our 27th straight quarter of profitability, and with the help of predictive modeling we are on track to maintain that record. This presentation is an examination of our lessons learned as part of our industry leading implementation of predictive modeling within commercial lines. We will be relating some poignant change management opportunities and other issues particularly significant for commercial lines insurance companies and related service providers to be aware of as they begin their own forays into applying predictive analytics within their organizations.

Moderator: Karl Rexer, Rexer Analytics

Speaker: Joel Appelbaum, Chief Analytics Officer, Zurich

Speaker: Steve VanDee, PMP, AVP of Underwriting Transformation, Zurich

3:55pm-4:45pm
Track 2: Product Recommendations
Room: Walnut A&B
Lessons That We Learned from the Netflix Prize

Netflix announced in Oct 2006 the 1 million dollar Netflix Prize competition to create a better recommender system, which just ended this summer. During the 3 years of the competition, many new approaches have been developed, and state of the art in the field of recommendation algorithms has been completely changed. In this session, we survey the most interesting algorithms of these 3 years, giving an insight into how new methods were developed by successively changing already existing ones.

Moderator: Eric Siegel, Program Chair, Predictive Analytics World

Speaker: Istvan Pilaszy, Prize Team, Gravity R&D


Delegates may choose to attend either session at this time.

4:45pm-5:30pm
Track 1: Insurance
Room: Magnolia
Case Study: Aflac
Establishing a Customer Retention Analytics Framework

Now more than ever, customer retention is a key performance indicator for many organizations. Unfortunately, assessing customer retention in any industry can prove to be quite a challenge. In this session we will review key steps for establishing and executing a customer retention analytics solution. A customer retention analytics solution should seek to define key metrics, gauge an organization's retention performance, identify/quantify sources of impact, and use predictive modeling to forecast future performance. Our modeling approach includes rigorous data-mining, regression analysis, and incorporation of additional assumptions for scenario modeling; each of which will be addressed. In addition, we will share our challenges and lessons learned along the way.

Moderator: Karl Rexer, Rexer Analytics

Speaker: Heather Avery, Manager, Business Analytics, Aflac

4:45pm-5:30pm
Track 2: Education
Room: Walnut A&B
Case Study: Walden University, Kendall College, University of Liverpool
The Use of Lead Scoring Solutions in the For-Profit Education Industry

The implementation of a lead scoring solution has the potential to render both Sales and Marketing organizations a competitive advantage. The case study presented will examine the use of lead scoring to (1) structure the on-line media buying process and (2) potential to drive contact management strategy with prospective students.

Moderator: Eric Siegel, Program Chair, Predictive Analytics World

Speaker: Christopher Scandlen, Laureate Higher Education Group

Speaker: Sherry Bennett-Flatt, Business Intel., Laureate Higher Edu. Group




Thursday October 22, 2009

Full-day Workshop
Room: Hickory
The Best and the Worst of Predictive Analytics:
Predictive Modeling Methods and Common Data Mining Mistakes

  • Workshop starts at 9:00am
  • First AM Break from 10:00 - 10:15
  • Second AM Break from 11:15 - 11:30
  • Lunch from 12:30 - 1:15pm
  • First PM Break: 2:00 - 2:15
  • Second PM Break: 3:15 - 3:30
  • Workshops ends at 4:30

Speaker: John F. Elder, Ph.D., CEO and Founder, Elder Research, Inc.


Register

Go to Top of Page


Linked In Linked In


PAW Highlights Video

2009 Sponsors















click here for sponsorship opportunities.

KDnuggets
Applied Forecasting
Training-Classes.com
BeyeNETWORK
CustomerThink Corp.

PRWeb

©2009 Predictive Analytics World
Produced by Prediction Impact, Inc. and Rising Media, Inc.

Predictive Analytics Company           Predictive Analytics Event Producer