Conference Day 1: Monday, September 30, 2013 |
8:00-9:00am • Room:
Commonwealth Hall
Registration & Networking Breakfast
9:00-9:45am • Room: Cityview
Keynote
The Prediction Effect, the Data Effect, and the Persuasion Effect
What are the underlying principles that make predictive analytics effective? Why is data predictive, why is imperfect prediction valuable, and what type of prediction succeeds to persuade? You have heard of the butterfly, Doppler, and placebo effects. In this session, PAW founder Eric Siegel covers the Prediction, Data, and Persuasion Effects. Each of these Effects encompasses the fun part of science and technology: an intuitive hook that reveals how it works and why it succeeds.
Attendee's receive a free copy of the related book by Eric Siegel, Predictive Analytics: The Power to Predict Who Will Click, Buy, Lie, or Die.
Speaker: Eric Siegel, Founding Chair, Predictive Analytics World
[ Top of this page ] [ Agenda overview ]
Diamond Sponsor Presentation
Measure Right, Manage Forward, Mobile Makes a Difference
Modern consumers are everywhere, all of the time. As this new generation of customer continues to evolve, with mobile customer experience as the change agent, so must the analytics used to measure the experiences they have with companies and organizations. ForeSee Senior Director of Mobile, Media and Entertainment, Eric Feinberg , will discuss Next Generation Customer Experience Analytics as a system of metrics that goes beyond single-number measurements and eliminates outdated metric silos to better support today's multi-channel, multi-device world we live in. He will explain what this new generation of predictive analytics needs to be and how it can help you create an analytics platform that allows you to measure right and manage forward.
10:00-10:30am • Room:
Commonwealth Hall
Exhibits & Morning Coffee Break
10:30-10:40am • Room: Harborview 1
Track 1
Gold Sponsor Presentation
Predictive Analytics 2.0 – Data Science for the Enterprise
Scaling your big data across hundreds of clusters may be a cool IT project, but scaling analytical and predictive applications to thousands of users can help unleash the hidden value of that data. Getting tangible benefits from vast amount of information requires relevant domain expertise so that we can formulate the right questions, an infrastructure capable of accessing multiple data sources on demand and preferably without replication, a dynamic data visualization layer accessible by business analysts for dimension free data exploration, and a statistical modeling environment that enables data scientists to construct relevant features, select best models, and enable predictive applications that can be deployed to thousands of users across the enterprise.
10:30-10:40am • Room: Cityview
Track 2
Gold Sponsor Presentation
5 Predictions about Predictive Analytics
Organizations have utilized predictive analytics to predict sales, customer behavior and many other important outcomes. In this session, I will share 5 predictions about the future of predictive analytics and their implications for predictive analytics professionals.
10:40-11:25am • Room: Harborview 1
Track 1: Customer Insights
Case Study: KeyBank
Making Key Business Decisions with Analytics to Better Serve Customers
KeyBank, one of the nation's largest bank-based financial services companies created a centralized Client Insights Center of Excellence leveraging analytics to deliver its customer relationship strategy and better serve clients. KeyBank learned that becoming more insights-driven supports personal touch strategy. David Bonalle of KeyBank will share:
- how creating centralized insights organization with strong senior management committed to being insights-driven can lead to success
- how businesses can leverage analytics to optimize business decisions at every level and better serve customers
- how to assemble a team for insights-driven approach
- examples of how KeyBank's sales & marketing team use analytics for various business functions
10:40-11:25am • Room: Cityview
Track 2: Thought Leadership
My Five Predictive Analytics Pet Peeves
Predictive Analytics (PA) has become increasingly mature as a technical discipline over the past decade in part because it stands on the shoulders of the related disciplines of data mining and machine learning. However, there are recurring themes that permeate discussion boards and conferences that have become my personal pet peeves. This talk examines five of them and why they matter to practitioners, including why we must have humility in how far data science and algorithms can take us, and the value of business objectives, measuring "success," and measuring "significance."
See also Dean's popular post-conference workshops
Speaker: Dean Abbott, President, Abbott Analytics, Inc.
[ Top of this page ] [ Agenda overview ]
11:30am-12:15pm • Room: Harborview 1
Track 1: Sales Strategy
Case Study: Paychex
Shaping Sales Strategy with Predictive Analytics
As we know, predictive modeling brings art AND science together. Without it, many strategic decisions are left to the "gut," leaving enormous opportunities in the age of big data. Paychex leveraged expertise in Predictive Analytics to add an empirical layer to sales strategy decisions. With the addition of models to predict likely sales units and establish a yardstick to measure sales value by zip code, sales management became statistically informed as they made decisions regarding quota setting, territory alignment and market expansion. This session will describe how Predictive Analytics at Paychex was thereby granted a seat at the strategic table.
Tom Kern, Risk Modeling Analyst, Paychex
11:30am-12:15pm • Room: Cityview
Track 2: Holistic Marking Applications
Case Study: AutoNation
How Predictive Analytics Can Drive Marketing Strategy
AutoNation, the largest automotive retailer in the country, started to adopt Predictive Analytics in 2008 and it is now becoming a driving force in defining and improving marketing strategies. Every marketing program has a Predictive Analytics framework integrated with ROI projections to identify customers eligible and to maximize incremental profit generated by programs given budget constraints. The evolution of Predictive Analytics carries on at AutoNation. Full exploit of targeting opportunities of the customer base as well as continuous marketing spending optimization are some of the next steps to move toward a holistic view across programs and channels.
Vikash Singh, Director, Marketing Analytics, AutoNation
[ Top of this page ] [ Agenda overview ]
12:15-1:15pm • Room:
Commonwealth Hall
Lunch
1:20-1:30pm • Room: Cityview
Plenary Session
Industry Trends: Highlights from the 2013 Data Miner Survey
In the spring of 2013, over a thousand analytic professionals from around the world participated in the 6th Rexer Analytics Data Miner Survey. In this PAW session, Karl Rexer will unveil the highlights of this year's survey results. Highlights will include:
- key algorithms
- challenges of Big Data Analytics, and steps being taken to overcome them
- trends in analytic computing environments & tools
- analytic project deployment
- job satisfaction
Speaker: Karl Rexer, President, Rexer Analytics
Keynote
Big Data Changes Everything? Not Really.
For one thing, data has always been big. Big is a relative concept and data has always been big relative to the computational power, storage capacity, and I/O bandwidth available to process it. Michael Berry now spends less time worrying about data size than he did in 1980. For another, data size as measured in bytes may or may not matter depending on what you want to do with it. If your problem can be expressed as a completely data parallel algorithm, you can process any amount of data in constant time simply by adding more processors and disks.
This session looks at various ways that size can be measured such as number of nodes and edges in a social network graph, number of records, number of bytes, or number of distinct outcomes, and how the importance of size varies by task. Michael will pay particular attention to the importance or unimportance of data size to predictive analytics and conclude that for this application, data is powerfully predictive, whether big or relatively small -- big data is no big deal.
Speaker: Michael Berry, Business Intelligence Director, TripAdvisor
[ Top of this page ] [ Agenda overview ]
Vendor Elevator Pitches
2:35-3:20pm • Room: Harborview 1
Track 1: Media Spend Analytics
Case Study: A Fortune 50 Company
Driving Media Buying Efficiencies Up 15-25%
Twenty-Ten demonstrated, with a Fortune 50 CPG client, how data could be used to determine their key areas of media spend, significantly increasing response rates and efficiency. After targeting their efforts to all moms, Twenty-Ten's client redefined their target segment as the Ultra Value Conscious Mom. Twenty-Ten then used data to determine the ideal media that contained the most optimal consumers who fit their new Ultra Value Conscious Mom target segment, driving a 15.7% efficiency increase for print media and a 20.4% efficiency increase for cable TV.
2:35-3:20pm • Room: Cityview
Track 2: Uplift ModelingCase Study: Subaru of America
Uplift Modeling and Beyond
How do you drive business results and incremental value from your marketing investments? What good is your complex, real-time predictive model for banner advertising if you cannot determine if the offer or ad exposure actually nudged the user to convert to the action that you wanted them to take? Any model can find internet user cookies that are likely to convert - but uplift is required if you want to identify users who need that extra push across the finish line. Partnering with Rocket Fuel, Carmichael Lynch Analytics team brought uplift to Subaru of America with some very measurable positive results.
Peter Amstutz, Analytic Strategist, Carmichael Lynch
[ Top of this page ] [ Agenda overview ]
3:20-3:50pm • Room: Commonwealth Hall
Break / Exhibits
3:50-4:35pm • Room: Harborview 1
Track 1: Industry Applications
How Sports-Based Predictive Analytics Will Transform the Business World
Sports have specific rules, boundaries, and structures that make analytic modeling easier than it is in the more complicated and fluid worlds of business. However, by learning how to take predictive models from these experimental fishbowls of sports to the wild and wooly world of business, you can make predictive analytics more relevant and important in the workplace. Based on 17 years of experience in fantasy sports and business analytics, this presentation will translate multiple sport-based predictive analytics models to business process optimization opportunities.
3:50-4:35pm • Room: Cityview
Track 2: Advanced Methods - Sales Lead ScoringCase Study: Hewlett-Packard
Combining Structured And Unstructured Data To Identify Opportunities With SMB Clients
Unified communication (UC) solutions helps small and medium businesses (SMB) get over communication latency and achieve mobility and connectivity of workforce cost effectively. HP has collaborated with its' partner to devise a "go-to-market" strategy to target the SMB space with their own UC solution. To support this effort, we use a unique analytical solution comprising of a mix of predictive models - value based segmentation and logistic regression on structured data to identify top customers and partners for this program and marry this prediction with unstructured product sentiment analysis to position the UC solution in the market more effectively across channels.
[ Top of this page ] [ Agenda overview ]
4:40-5:30pm • Room: Harborview 1
Track 1: Healthcare Analytics
Case Study: New Directions Behavioral Health
Deploying Predictive Models In Virgin Waters: Predicting Behavioral Health Readmissions
Deploying predictive modeling in an organization for the first time can be difficult. This is especially true in industries like behavioral healthcare that are driven more by anecdotes than data. Getting management buy-in, convincing skeptics and producing a finished product with tangible results can be a long and trying road. However, a well thought out plan executed with precision can lead an organization skeptical of predictive modeling to embracing it. Fred Grunwald will discuss the steps, from beginning to end, of how a project to predict inpatient readmissions drove New Directions Behavioral Health to leverage and embrace predictive modeling.
Track 2: Customer SatisfactionCase Study: Analysis of Public Complaint Data
Ditch the Crystal Ball: Leveraging Predictive Analytics to Eliminate Customer Dissatisfaction Before it Spells Disaster for Your Business
As the amount of collected customer data continues to rapidly increase, many businesses are asking themselves, "How can I use this information to improve customer experience?" Based on continual in-depth analyses of the Consumer Financial Protection Bureau's consumer complaint database, Steven Ramirez, CEO of Beyond the Arc, will discuss:
- how to utilize big data to predict customer issues
- the risks and threats of ignoring customer complaints
- best practices for analyzing social media and customer feedback data to identify and address potential issues
Speaker: Steven Ramirez, CEO, Beyond the Arc
[ Top of this page ] [ Agenda overview ]
5:30-7:00pm • Room:
Commonwealth Hall
Reception / Exhibits
Boston Predictive Analytics MeetUp
Lightning Talks: "Product Launch Challenge, Next Generation Social Websites, Drug Adoption in Physician Networks, Brand Impact from Text"
The goal of the Meetup group is to help the local community further it's understanding and proficiency regarding Predictive Analytics through informative lectures, hands-on tutorials, and networking events. Our group has three main focal points: business applications, advanced mathematics, and computer science. Past events have included sentiment analysis, web content recommendations, social media and network analysis; interactive visualizations, as well as several events pertaining to the Big Data / Hadoop ecosystem.
Boston's Meetup Community: John Verostek
Building Data Science Teams: David Dietrich
dunnhumby Product Launch Challenge: William Li, George Tucker
Analytics for Next-Generation Social Websites: John Bottoms
Predicting the Adoption of Drugs in Networks of Physicians: Aaron Merlob
Discovering Brand Impact from Text: Catherine Havasi
www.meetup.com/Boston-Predictive-Analytics
Conference Day 2: Tuesday, October 1, 2013 |
8:00-9:00am • Room:
Commonwealth Hall
Registration & Networking Breakfast
Conference Chair Welcome Remarks
9:05am-9:25am • Room: Cityview
Platinum Sponsor Presentation
Predictive Coding - Document Review in Legal Matters and Investigations
As the volume of electronically stored information increases at an exponential rate, many organizations are looking for new ways to manage document review and rein in escalating costs. Meanwhile, advanced analytics techniques such as predictive coding are opening the door to new opportunities for corporate legal departments, government agencies, and outside counsel looking to make sense of this growing mountain of information. This presentation explores the use of predictive coding to allow for intelligent and faster decision making while achieving cost savings.
Keynote
Using Analytics to Power the New Style of IT
Session description is coming soon!
[ Top of this page ] [ Agenda overview ]
10:15-10:40am • Room:
Commonwealth Hall
Exhibits & Morning Coffee Break
[ Top of this page ] [ Agenda overview ]
10:40-10:50am • Room: Harborview 1
Track 1
Gold Sponsor Presentation
Addressing Privacy Concerns: Critical Features for Predictive Analytics Platforms
For many years, the primary focus of statisticians, mathematicians, data scientists, and data miners has been to produce accurate models and predictions. The flip side of success in this endeavor is that customers, patients, or others who are accurately "predicted" may--and often will--consider this to be an invasion of their privacy. The more effective the modeling tools have become, the more vocal are the critics of their un-"regulated" applications.
Governance of models and data (big or small) is a neglected topic in the discussions among predictive modelers, and it is an important driver of software requirements often missing from projects to implement predictive modeling platforms.
This session will discuss some key features of the STATISTICA Enterprise Decisioning Platform® that have made it a favorite in highly regulated industries. Given the recent coverage of invasion-of-privacy concerns due to exhaustive and effective data mining, it is critical that every organization has clear rules on who can access specific data, who approves models, who uses models, when to predict what features, and what documentation is to be produced to demonstrate that the application of models is fair. Enterprise analytics platforms must support features allowing the implementation of these policies and rules.
10:40-10:50am • Room: Cityview
Track 2
Gold Sponsor Presentation
Predictive Analytics 2.0 – Data Science for the Enterprise
Scaling your big data across hundreds of clusters may be a cool IT project, but scaling analytical and predictive applications to thousands of users can help unleash the hidden value of that data. Getting tangible benefits from vast amount of information requires relevant domain expertise so that we can formulate the right questions, an infrastructure capable of accessing multiple data sources on demand and preferably without replication, a dynamic data visualization layer accessible by business analysts for dimension free data exploration, and a statistical modeling environment that enables data scientists to construct relevant features, select best models, and enable predictive applications that can be deployed to thousands of users across the enterprise.
10:50-11:10am • Room: Harborview 1
Track 1: Analytics Talent
Using Analytics to Build Your Analytics Bench: Announcing 2012 Analytics Professionals Study Results
Many innovative businesses and IT organizations appreciate the competitive advantage analytics capabilities can provide and have ambitions to reach increasing levels of analytics maturity. However, the well-documented shortage of analytic talent leaves many firms without a strong analytic talent bench and little knowledge about how and where to find analytics professionals needed to get there. In this presentation, Greta Roberts will discuss results from a major 2012 Study of Analytics Professionals that crosses industries, experience and skills. Practical insights shared include key best practices, trends and correlations that lend unexpected insight into building a strong and scalable analytic talent bench.
Speaker: Greta Roberts, Faculty Member, International Institute for Analytics
11:15-11:35am • Room: Harborview 1
Track 1: Internet Security
Case Study: The Bill and Melinda Gates Foundation
Predicting Threats For The Gates Foundation - Protecting Our People, Investment, Reputation and Infrastructure
Chris Sailer will discuss how The Gates Foundation exploits Predictive Analytics to ensure that its global mission is advanced safely and securely. This analytical capability provides a 360 view of known risks and identifies emerging risks. Our threat scoring system collects and analyzes unsolicited inbound correspondence providing real-time diagnostic intelligence. We synthesize a factor and motive-based conceptual model with behavioral modeling to operationalize threat prediction. If a correspondence is classified as high threat - it is automatically routed to The Foundation's protective intelligence specialists. They utilize Saffron to determine root cause to take appropriate action.
10:50-11:35pm • Room: Cityview
Track 2: Insurance – Risk Modeling
Case Study: Opta Information Intelligence
Determining True Non Linear Variable Relationships and Enhancement of Property Insurance Risk Models
Richard Boire will discuss how Building Property Insurance Risk Models was a relatively new area of focus for Boire Filler Group. During the model discovery stage, a number of non linear patterns emerged from our variable and exploratory data analysis. Our challenge was to determine whether these relationships were indeed noise or in fact represented viable business patterns that needed to be captured in the model. Vetting our results with key domain business experts, we were able to transform these variables to better reflect the non-linear relationship. These transformations yielded significant model improvement with an ROI increase of 50%.
[ Top of this page ] [ Agenda overview ]
11:40am-12:00pm • Room: Harborview 1
Track 1: Online Ad Optimization
Case Study: Quantcast
Real-Time Bidding: A New Era in Advertising
For decades, cost per thousand (CPM) pricing controlled how all advertising was bought and sold, and how media properties were compared. While a focus on CPM-based buying provided a foundation for the first decade of online display growth, it is now preventing the online advertising industry from realizing the full potential of its biggest game changers "real-time bidding" (RTB). Today, RTB presents a huge opportunity for buyers to "take it to the next level." Focused on value, buyers can evolve from haggling over costs to collaborating with media partners to find strategic value and ROI they are both committed to delivering.
Speaker: Michael Recce, Vice President of Modeling, Quantcast
12:05-12:25pm • Room: Harborview 1
Track 1: Lifetime Value
Case Study: AOL
How Much Are You Worth? - Calculating Customer Lifetime Value
How can we accurately measure how much someone who comes to your site will be worth? In order to maximize ROI and LTV, Brett Cohen will discuss how AOL Search took a step back from the complex LTV models and created a 3 pronged model that takes audience, engagement (and churn), and monetization into account to measure the value of users coming in from different properties. See how AOL visualizes this data and how they cut the time to decisions by over 80%; allowing the company to divest from ROI negative partnerships nearly immediately, as well as invest more with partners who have optimal performance.
11:40-12:25pm • Room: Cityview
Track 2: Fraud Detection; Analytics Project Management
The Organic Evolution of a Data Analytics Team
The USPS Office of the Inspector General CAPE Team supports data analytics for hundreds of employees covering multiple program areas. But just three years ago, they were a small group enlisting outside help to examine Contract Fraud. With success, the team, its mission, and its services have grown organically, each step determining the direction of the next. Sarah Will, a Data Scientist with Elder Research, will discuss this natural progression, with specific examples around data, modeling, and visualization, where the iterative development created a self-sustaining productive environment.
[ Top of this page ] [ Agenda overview ]
12:25-1:25pm • Room: Commonwealth Hall
Lunch
[ Top of this page ] [ Agenda overview ]
Special Plenary Session
General Lessons We Can Learn From Blackbox Trading
Beating the market with skill, rather than luck, is so hard that it's arguably impossible. A strong working approximation is that markets are efficient - that prices reflect available information almost instantaneously. Accordingly, we have failed often. But our success building quantitative investment systems has been great - most notably with a hedge fund that beat the S&P-500 every year for a decade, with only 2/3rds the risk (volatility). Dr. John Elder will highlight key lessons learned from the long battle, and how those insights have helped solve many other predictive analytics challenges.
See also John's popular post-conference workshop
2:15-2:30pm• Room: Cityview
Vendor Elevator Pitches
[ Top of this page ] [ Agenda overview ]
Expert Panel
Big Data for Predictive Analytics
If Big Data begs the question, "What to do with all this data?" Predictive Analytics answers, "Learn from it to predict behavior." But just how much predictive payoff comes with going so big? This expert panel will address the new demands on Predictive Analytics solutions and best practices as data grows to enormity, and will recommend tactics to fully leverage data's growing magnitude to improve the business performance of Predictive Analytics initiatives.
Satish Lalchand, Deloitte Financial Advisory Services LLP
Eric Feinberg, Senior Director of Mobile, Media and Entertainment, ForeSee
3:15-3:45pm • Room: Commonwealth Hall
Exhibits & Afternoon Break
[ Top of this page ] [ Agenda overview ]
3:50-4:35pm • Room: Harborview 1
Track 1: Energy Management
Case Study: UCSD - San Diego Supercomputer Center
Predictive Analytics for Smart Grids
As demand for cost-effective energy and resource management continues to grow, intelligent automated building solutions are necessary to reduce energy consumption, increase alternative energy sources, reduce operational costs and find interoperable solutions that integrate with legacy equipment without massive investments in new equipment and tools. UC San Diego's 1200-acre campus provides a forward-thinking, innovation engine to improve operational efficiency, lower operating costs, and reduce the overall carbon footprint of its 45 MW peak load Smart Grid. In collaborations with the commercial sector, PACE is developing a scalable decision support system for large scale, predictive analytics for intelligent real-time energy management.
3:50-4:35pm • Room: Cityview
Track 2: Persuasion Modeling (aka Uplift Modeling)
Case Study: Obama for America
Pinpointing the Persuadables: Convincing the Right Voters to Support Barack Obama
Prior to President Obama's reelection campaign, standard practices for persuading voters—that is, changing their minds—were unscientific and driven by long-standing assumptions and hunches. Campaigns targeted broad categories of typically "independent" voters and assumed that these voters would respond to a persuasive message. That all changed with the Obama reelection. Campaign leadership knew that 2012 would be different from 2008. Turning out likely supporters was not enough; the campaign had to persuade voters that President Obama was a better choice than Mitt Romney. Daniel Porter, Director of Statistical Modeling for the Obama Campaign, will discuss how his team used the results from a large-scale randomized, controlled experiment to model which individual voters were most likely to be persuaded, and how this model served as the basis for targeting decisions across many aspects of campaign.
[ Top of this page ] [ Agenda overview ]
4:40-5:30pm • Room: Harborview 1
Track 1: Big Data Quality
Case Study: Citibank
The Importance of Data Quality in the Context of Big Data
Many businesses and institutions are pursuing the use of Big Data as a means of better understanding the current state as well as improving predictions. It is important to understand that the strength of those predictions is based almost entirely on your understanding of the data and its inherent quality.
This presentation will focus on the data quality framework as it pertains to Big Data related predictive analytics with some appropriate examples.
Bob Granese, VP, Data Analyst, Data Quality Analytics & Improvement, Citigroup
Jagmeet Singh, Citigroup
4:40-5:30pm • Room: Cityview
Track 2: Online Ad Optimization
Case Study: CompassLabs
Predictive Analytics in Social Media Advertising
The new wave of social media (Facebook, Twitter, etc.) in the last decade has made it easier than ever for marketers to reach the right customers at the right time with the right products and offers. However, the traditional analytical tools and methodologies are often insufficient due to the rapidly growing volumes of data as well as increasing importance of analyzing textual and unstructured data in this space. Mahesh Kumar will present a case study on applying data analytics to help a leading credit card company improve new customer acquisitions through Facebook campaigns.
[ Top of this page ] [ Agenda overview ]
