Tuesday, April 24, 2012
Instructor: John Elder, CEO & Founder, Elder Research, Inc.
Wednesday, April 25, 20127:30am - 7:30pm
Exhibit Hall Open
Registration & Breakfast
Persuasion by the Numbers: Optimize Marketing Influence by Predicting It
Speaker: Eric Siegel, Program Chair, Predictive Analytics World
Lightning Round of 2-Minute Sponsor Presentations
Break / Exhibits
Case Study: MTV Networks
Predictive Social Marketing – Sentiment Forecasting and Impact on Success
Every summer, music fans worldwide look forward to one of the biggest music events of the year—the MTV Video Music Awards (VMAs). This year's VMAs turned out to be one of the world's largest, simultaneous social viewing experiences ever. Leading up to this year's VMAs, MTV marketers set out to grow the brand's social media presence and drive awareness. With 85 million MTV Facebook fans and more than three million Twitter followers—the stage was set for a firestorm of conversation and sharing. In order to assist MTV in their primary goal of gaining deeper insights into relationships between social activity and engagement with digital content, we combined Twitter and MTV.com data streams with text mining and predictive analytics techniques. As marketing continues to develop, sentiment forecasting will be critical to success in optimizing published content for both publishers and advertisers.
Special Featured Session
Multiple Case Studies: Disney, HSBC, Pfizer, and others
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. 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:
- Cross-selling for HSBC
- Biometric identification for Lumidigm (for Disney)
- Optimal decisioning for a leading high-tech retailer
- Quick decisions for the Social Security Administration
- 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 & Up John (now Pfizer)
Speaker: John Elder, CEO & Founder, Elder Research, Inc.
Platinum Sponsor Presentation
Managing Forward: Analytics For Today's Multi-Channel, Multi-Device Consumer
When done right, customer satisfaction measurement can yield more than just insights into how well your company, brand, or channel (e.g., web, mobile, store) is performing today. It can also predict the likelihood of customers to engage in critical future behaviors. However, not all methodologies are created equal. They must answer three essential questions of management while demonstrating success not only in theory but in the marketplace.
Speaker: Larry Freed, President & CEO, ForeSee
Lunch / Exhibits
The Five Myths of Predictive Analytics
Predictive Analytics is powerful, it can help you predict an event or a behavior at a an individual customer level. It can help you spot golden nuggets from the deep-wide-big data ocean; But is also one of the techniques which is not very well understood. With all the recent buzz about Predictive Analytics, it does seems like a new technique in the tool box. Is that so? In this keynote, we will ground ourselves in the reality of building and maintaining an impactful Predictive Model and explore questions like
- Is Predictive Analytics new?
- Is it a crystal ball?
- Is it perfect?
- Can it be built quickly and cheaply?
- Is it going to solve all my business problems?
- Does it always work?
- Can anybody learn how to build a model?
Speaker: Piyanka Jain, CEO, Aryng.com, Former PayPal Business Analytics Head
Case Study: CIBC
Creating Value Segmentation to Drive Business Strategies
Segmentation is used in a variety of ways to help organizations understand customer behavior and tailor their services. This case study presents how CIBC built a value based segmentation that focuses on customer's current and potential value to identify business opportunities and risks. Adoption and driving change in business strategy and processes are also covered. The concepts involved are not limited to banking and should be generally applicable to all industries.
Speaker: Daymond Ling, Senior Director of Modelling & Analytics, CIBC
Case Study: Capital One Financial
Competitive Training of Predictive Modelers
Competition between teams for prizes and awards is a well known mechanism for unleashing the creative forces of predictive modelers to develop new modeling techniques, methodologies, and approaches - and, of course, build better models. The Netflix prize competition is a recent example of the power of such competition and coopetition. Capital One Canada set up a competition amongst students of predictive modeling at the University of Waterloo to see what they could do on the kinds of problems typically faced by a lending institution. The format was an integrated case study involving business strategy development around a core predictive model. Perhaps not surprisingly, student-built models didn't outperform models built in-house, but the students' creative approaches and outside-the-box thinking were truly inspiring. Most importantly however, the demand by students and faculty to work on real problems and data sets was almost overwhelming and the feedback by students about the importance of this competition for learning about predictive modeling was surprising. Since student learning is the primary value of this kind of competition, we suggest that multiple institutions could combine efforts to create excitement about predictive modeling and establish a mutually beneficial relationship between academia and industry.
Break / Exhibits
Top 10 Data Mining Business Mistakes
Most analytics talks are technical – describing algorithms, data management, software options, etc. – that best extract value from data. And, great technology helps, in our experience, over 90% of projects to meet their technical goals. However, only about 65% of solutions seem to actually be deployed at the client organization. Astonishingly then, business risk has proven far greater than technical risk as an obstacle to realizing the huge ROI possible from predictive analytics.
This talk focuses on the business pitfalls of managing a data mining engagement, complementing John Elder's popular technical chapter on Top 10 Data Mining Mistakes (also covered during his pre-PAW workshop). We address organizational and management mistakes commonly made by either the client or the consulting firm, and illustrate select ones with real-world examples. Anyone who is considering or actively engaged in mining data will benefit from these cautionary tales!
Speaker: Jeff Deal, Vice President of Operations, Elder Research, Inc.
Case Study: Crawford & Company
Predictive Analytics in Property Insurance Claims: Findings and Lessons Learned
To gain advantage in the highly competitive insurance market, property insurers are increasingly looking at using predictive analytics to optimize claim processing. However, achieving sustainable value has not been easy because of variety of reasons. If used properly, predictive analytics in property claims can provide tremendous cost saving and quality improvement. We mined property claims 'descriptive, transactional, and unstructured data' to determine what works, what does not, how to use predictive models, what challenges to expect, and what benefits to expect. This presentation is a case study that walks the audience through the objectives, methods, findings, and lessons learned.
Reception / Exhibits
Thursday, April 26, 20127:30am - 7:30pm
Exhibit Hall Open
Registration & Breakfast
Open Question Answering
Putting IBM Watson to Work
IBM's Watson captured the imagination of millions when it beat the all time champions of the US game show, Jeopardy!. To do so, it overcame traditional limitations of computers by communicating in natural human language, churning through 200 million pages of unstructured data to find answers in three seconds, and learning from each experience to improve performance over time. But as impressive as this accomplishment was, it was only the beginning. IBM is working closely with leading organizations in a variety of industries to put Watson to work. The possibilities are endless! Join Edward Nazarko, a leading IBM Architect, in an engaging discussion of ways that Watson is using predictive models to revolutionize expectations of how computers can help organizations in all industries live and work better.
Speaker: Edward Nazarko, Client Technical Advisor, IBM
Break / Exhibits
Case Study: U.S. Special Forces
Hiring and Selecting Key Personnel Using Predictive Analytics
Hiring and selection of personnel in specialized work environments incurs huge direct and opportunity costs for organizations. One of the largest challenges is that the selection process is often left in the hands of those with either high experience in the domain area but little experience in selection or vice versa. Predictive Analytics and statistics can play a critical role in formalizing and automating much of the selection process. This session provides an overview of the selection processes using both measures of skills and psychological measures to quantify IQ, domain knowledge, grit, and determination. Examples will be drawn from hiring practices for Special Forces (such as Army Rangers and Navy SEALs) and predictive analytics teams.
Speaker: Dean Abbott, President, Abbott Analytics, Inc.
Case Study: Alberta Motor Association
Insurance Pricing Models using Predictive Analytics
The use of predictive analytics solutions as a pricing tool for insurance is a very recent phenomenon amongst actuaries. This case study examines what tools were used in the past and what has led to the adoption of predictive analytics solutions within the actuarial discipline. Particular emphasis is devoted to the significant data challenges which are unique to the insurance pricing sector. At the same time, attendees will learn the process that was adopted in building these tools. More importantly, attendees will understand how to demonstrate the value or benefit of predictive analytics solutions over existing actuarial tools.
Speaker: Richard Boire, Partner, Boire Filler Group
Lunch / Exhibits
Wise Enterprise: Best Practices for Managing Predictive Analytics
Your company is trigger-happy for predictive analytics, and there's plenty of excitement, momentum and public case studies fueling the flames. Are you destined for success or disappointment? Is it a sure-fire win to gain buy-in for a promising analytics initiative, equip your most talented practitioners with a leading solution, and pull the trigger?
This panel of leading experts will address the holistic view. What are the most poignant and telling failures in the repertoire, and where is the remedy? Beyond the management of individual analytics projects, what enterprise-wide communication processes and other best processes provide best security against project pitfalls? Stay tuned for big answers to these big questions.
Case Study: IBM
If We Host It; Will They Come? Predictive Modeling for Event Marketing
Every year IBM (like all large companies) spends millions of dollars on hundreds of Marketing events, from conference participation through seminars to briefing and education events. The 'gold' questions for Marketing are: 1) Are we engaging in the right events, 2) Which events are likely to bring in the most success (measured in terms of revenue), and; 3) Given a finite Marketing budget, how should we spend that budget to optimize the value we are receiving. To answer these questions, IBM's Event Marketing engaged in an exercise developing predictive models to enhance their insight around events activities. This presentation will review the approach, predictive models and will offer some next steps discussion looking into the future of Event Marketing predictive model development.
Break / Exhibits
How do you produce estimates in the presence of ambiguous, spotty, or unreliable data? This case study describes a tool intended to do just that, where the particular goal of the tool was to estimate the cost of developing military grade software applications. Instead of relying on an expert system to produce estimates, the tool uses many relatively naive regression models and other relatively simple off the shelf statistical techniques. Each model is tested against a large repository of projects and assigned a score based on its accuracy forecasting outcomes of projects most similar to the target project. The estimates generated by well-performing models are combined into a weighted estimate for the target. The same tool has since been used to estimate costs associated with military grade hardware.
Speaker: Mike Kimel, Principal Consultant, Analytic Economics
Predictive modeling has long been applied to target customers likely to purchase a specific product; by understanding the characteristics of those who have purchased in the past, you can predict the future. However, what happens when you don't have history? When the product you are offering is new in the market. This case study shows how DiG helped a client face this challenge. Leveraging past modeling efforts around churn and upsell, and sprinkling in some business knowledge, DiG developed an Engagement Index which helps the client identify those customers who are engaged and thus likely to be open to new offerings from the company. This case study will also show how this index was used in conjunction with a client value model to understand where additional opportunities and risks lay.
Case Study: Verizon Wireless & BHP Billiton (Australian mining company)
Visualizing Forecast Models with Interactive Scenario Analysis to Optimize Profitability
A $20B pharmaceutical company spends more than $5B annually on new products. Planning and modeling teams updated staffing and projections but the company still had difficulty hitting targets. Earnings are flat and the share price languished. Visualizations enabled the all the stakeholders to see some of the underlying issues. Revenue and share price are now growing again.
Visualization is a key tool for the analyst to organize and assess data, understand patterns to help build models and assess results of predictive models. Learn what some of the tools and techniques are in this session showing examples of visualization and predictive analytics:
A. A top 3 international resource company actively visualizes scenarios across key performance metrics to steer the corporation through strategic decisions.
B. A $20B pharmaceutical company visually combines multiple short and long-term projections to streamline their forecast models, optimize decisions and increase revenue.
C. A top 5 wireless organization is able to combine prediction with what-if analysis and visualization to rapidly assess, evaluate alternatives and implement changes in minutes to competitive challenges to any customer/product segmentation.
Speaker: Richard Brath, Partner, Oculus
Workshop sponsored by:
Friday, April 27, 2012
Instructor: Dean Abbott, President, Abbott Analytics, Inc.