September 16th 2014

Wise Practitioner – Predictive Analytics Interview Series: Marty Kohn, M.D. of Jointly Health

Wise Practitioner – Predictive Analytics Interview Series: Marty Kohn, M.D. of Jointly Health

By: Jeff Deal, Program Chair, Predictive Analytics World Healthcare

In anticipation of his upcoming conference keynote at Predictive Analytics World Healthcare in Boston, “Big Data and Clinical Decision Support,” we asked Marty Kohn, M.D., Chief Medical Scientist at Jointly Health, a few questions about his work in predictive analytics.

Q: In your work with predictive analytics, what area of healthcare are you focused on (i.e., clinical outcomes, insurance, quality improvement, etc.)?

A: Jointly Health focuses on patients with complex chronic diseases to improve health, reduce avoidable hospitalizations and acute care events and, as a result of decreased need for expense acute care, reduce costs.

Q: What outcomes do your models predict?

A: We predict which patients are likely to deteriorate so that a timely intervention can avoid the problem.

Q: How does predictive analytics deliver value at your organization? What is one specific way in which it actively drives decisions?

A: By identifying patterns in home monitoring physiologic data, coupled with interaction with the patient and the patient’s caregivers, we can give the care team early warning of a worsening of the patient’s clinical status. We develop such patterns in a way that is unique for each patient, allowing the care team sufficient warning to treat the problem when it is more likely to be successful.

Click here to read the rest of this interview. 

No Comments yet »

September 9th 2014

Wise Practitioner – Predictive Analytics Interview Series: John Foreman of MailChimp

Wise Practitioner – Predictive Analytics Interview Series: John Foreman of MailChimp

By: Eric Siegel, Founder, Predictive Analytics World

In anticipation of his upcoming conference keynote at Predictive Analytics World Boston, “Problems, then Techniques, then Toys. Keeping Your Predictive Analytics Right-side Up,” we asked John Foreman, Chief Scientist at MailChimp, a few questions about his work in predictive analytics.

Q: In your work with predictive analytics, what behavior do your models predict?

A: At MailChimp, we use predictive modeling across the application to improve the experiences of our users. Some examples:

  • We predict users who are unlikely to send spam, and we allow them to begin sending email through the system without manual account vetting (manual vetting slows people down by a day)
  • We predict users who are likely to send spam, and we shut them down before they send in order to protect our email-sending ecosystem
  • We predict users who are on a free account but who are likely to pay in the future. We then give them the same customer support given to currently paid users
  • We predict users who are most certainly not bots and we remove reCAPTCHA entirely from the app for them
  • We predict the knowledge base articles that a user is most likely interested in when they contact customer support
  • We predict the best time to send an email address marketing content and provide that to users in our Send Time Optimization (STO) system
  • Given a small segment of email addresses, we predict other email addresses on a user’s list that have the same interests to facilitate better segmentation and targeting
  • We predict demographic data on email addresses

These are just some examples of the different models in play.

Q: How does predictive analytics deliver value at your organization? What is one specific way in which it actively drives decisions?

A: Predictive analytics is a key part of our user on-boarding and compliance process. MailChimp has over 6 million customers, and without predictive modeling, the company would be left linearly scaling the headcount of customer support and compliance. Predictive models enable us to automate the easy jobs, allowing our compliance personnel to hunt down the worst of worst in terms of bad actors. This lowers our headcount, saving us a great deal of money. We are able to manage 6 million customers with less than 300 people total at the company.

Furthermore, our user-facing predictive products (Send Time Optimization & Segment

Click here to read the rest of this interview.

 

No Comments yet »

September 2nd 2014

Wise Practitioner – Predictive Analytics Interview Series: Jack Levis of UPS

Wise Practitioner – Predictive Analytics Interview Series:

Jack Levis of UPS

By: Eric Siegel, Founder, Predictive Analytics World

In anticipation of his upcoming conference keynote at Predictive Analytics World Boston, “UPS Analytics – The Road to Optimization,” we asked Jack Levis – Senior Director, Process Management at UPS, a few questions about his work in predictive analytics.

Q: In your work with predictive analytics, what behavior do your models predict?

A: We use a tremendous number of predictive and prescriptive models at UPS. They are used to help make decisions, which range from where to build a facility and what type of aircraft to purchase to which packages go in each trailer and how to maintain our delivery fleet.

We currently have 700 dedicated resources working on a system called ORION, which has been called “arguably the world’s largest Operations Research Project.” With ORION, we are using analytics to determine the best way for a driver to serve our customers at the lowest cost.

Q: How does predictive analytics deliver value at your organization? What is one specific way in which it actively drives decisions?

A: We do not do anything by the “seat of our pants.” Analytics is engrained so deep in our culture, it is difficult to separate analytics driven decisions from normal business processes.

In 1954, our CEO said, “If we did not have operations research, our rate of growth might have been affected. As we grow in size, our problems increase geometrically. Without Operations Research, we would be analyzing our problems intuitively only, and we would miss many opportunities to get maximum efficiency out of our operations.”

Analytics has helped UPS make better decisions in all parts of our business.

Q: Can you describe a successful result, such as the predictive lift of your model or the ROI of an analytics initiative?

A: In 2003, UPS began using predictive models to better plan our delivery operations. This suite of tools called Package Flow Technologies along with Telematics has been responsible for a yearly reduction of 85 million miles driven per year. This reduced our fuel needs by over 8 million gallons and reduced carbon emissions by 8,500 metric tons.

In addition, because the analytics and business processes are fully aligned we have been able to deploy new products for customers. UPS’ MyChoice is a prime example of that.

Click here to read the rest of this interview.

No Comments yet »

August 26th 2014

Wise Practitioner – Predictive Analytics Interview Series: Sameer Chopra of Orbitz

Wise Practitioner – Predictive Analytics Interview Series: Sameer Chopra of Orbitz

By: Eric Siegel, Founder, Predictive Analytics World

In anticipation of his upcoming conference keynote at Predictive Analytics World Boston, “Blackjack Analytics: A Surprising Teacher from Which All Businesses Can Learn,” we asked Sameer Chopra, GVP of Advanced Analytics, Orbitz Worldwide, a few questions about his work in predictive analytics.

Q: In your work with predictive analytics, what behavior do your models predict?

A: We have predictive models for various applications. For instance:

  • In the CPC (cost-per-click) online marketing channels we have response models to predict the Revenue Per Click (RPC).
  • We have user level purchase-propensity models (i.e. the likelihood of Eric transacting on Orbitz in say the next 24 hours – vs. Sameer transacting). As you can imagine, this can be an effective lever to use real-estate on our site effectively (eg: whom to show ads to, as a simple use case).
  • We also have models to predict user-attrition. Unlike industries with black & white subscriber models, we live in the gray — so a tool like Survival analysis can be a helpful friend.
  • We also have built credit-card fraud models.
  • I should highlight that we do actively test uplift models to get smarter about segments to pursue and whom not to. etc.

These are just some examples of the different models in play.

Q: How does predictive analytics deliver value at your organization? What is one specific way in which it actively drives decisions?

A: There are numerous areas where PA is driving value. One specific example where it actively drives decisions is in online marketing: determining how much to bid in CPC channels such as Google Adwords, Google Hotel Price Ads (HPA), or Travel Research partners such as TripAdvisor etc.

Click here to read the rest of this interview.

No Comments yet »

August 20th 2014

Top Reasons to Attend Predictive Analytics World for Government

Eight Reasons to Attend Predictive Analytics World for Government

The clock is ticking and Predictive Analytics World for Government kicks off in less than ONE month on September 15th in Washington D.C. Have you been thinking about registering? Here are just some of the reasons you should attend this year’s conference.

  1. Stay on the Cutting Edge of Predictive Analytics in Government – Predictive Analytics is growing in the government and public sectors. This conference is designed to help agency managers understand how they can apply predictive analytics to more effectively and efficiently accomplish their mission. Learn about the latest progress in the field and apply it to your work.
     
  2. Solve Real-World Problems – More and more, data analytics are being used by government agencies to solve real-world problems. They are utilized to solve issues in surveillance, security, fraud detection, healthcare and in many other realms of the government. Learn about the difference that predictive analytics can make in these scenarios and in many more.
     
  3. Hear from Top Keynotes and Speakers – With speakers from the IRS, Federal News Radio, Elder Research Inc. and, potentially, the Senate, you will learn important information from authoritative leaders in the predictive analytics and government world.

 

  1. Digest the Incredible Agenda – The 4th annual Predictive Analytics World for Government agenda is packed with substantial speakers, workshops, topics, and content that you won’t want to miss. View the full agenda here.
     
  2. Exchange Best Practices – Witness many case studies and hear experiences that have paved the way toward the best implementation of predictive analytics in government. Discover similar routes to apply toward your company’s future success.
     
  3. Build your Network with Skilled Executives in your Field – Join 400+ participants at PAWGov. With hundreds of fellow federal, state, and local government leaders, analysts, and program managers, you will build a long-lasting network that grows far beyond the conference walls.
     
  4. Be Part of the ONLY Vendor-Neutral Predictive Analytics Conference for Government – With such a specific focus, this conference will nail down exactly what the hot topics are relating to your work with predictive analytics among government.
     
  5. Increase Your Performance – The conference delivers case studies, expertise and resources to empower you in federal, state, and local government to best achieve your objectives. Apply the takeaways you’ve gained immediately to increase your performance and deliver more effective and efficient results.

 

This is just a glimpse of the many reasons to attend Predictive Analytics World for Government. Join us for September 15th – 18th at the Grand Hyatt Washington. We look forward to seeing you there!

 

No Comments yet »

July 22nd 2014

Eleven Analytics Conferences – New Verticals

                                                 

Eleven Analytics Conferences – New Verticals
 
Here is the line-up of Predictive Analytics World conferences and workshops coming over the next year. I'm the founding chair, so please let me know if you have any questions about any event programs.
 
Predictive Analytics World covers all the basics for both expert practitioners as well as newcomers. As the universal, cross-vendor meeting place that brings together the who's who of predictive analytics, PAW delivers not only unique opportunities to gain knowledge, but the industry's premier networking event.
 
NEW VERTICALS: HEALTHCARE & WORKFORCE. In addition to the usual business-focused PAWs, and the annual PAW Government, the inaugural PAW Healthcare (http://pawcon.com/health) & PAW Workforce (http://www.pawcon.com/workforce) are coming – these arenas are exploding with movement and interest.
 
CALL FOR SPEAKERS now open for 2015 events – see http://pawcon.com/cfs
 
———————–
 
2014 EVENTS – IN COMING MONTHS:
 
PAW GOVERNMENT – Sept 15-18, 2014
Discount Code for $150 off: LIN150
Early Bird Registration Ends July 25
 
PAW BOSTON – Oct 5-9, 2014
Discount Code for $150 off: LIN150
Early Bird Registration Ends Aug 15
 
PAW HEALTHCARE – Oct 6-7, 2014
Early Bird Registration Ends Aug 15
 
PAW LONDON – Oct 29-30, 2014
Discount Code for $150 off: LIN100
Early Bird Registration Ends Sept 1
 
PAW BERLIN – Nov 4-5, 2014
Discount Code for $150 off: LIN100
Early Bird Registration Ends Sept 20
 
WORKSHOPS: A plethora of 1-day analytics workshops are held alongside PAW
(For other cities, navigate from: http://www.pawcon.com)
 
- – - – - – - – - – - – -
 
2015 EVENTS – SAVE-THE-DATES / CALLS-FOR-SPEAKERS:
 
PAW SAN FRANCISCO – March 29-April 2, 2015
Speaker proposals due Sept 26 – http://pawcon.com/submit.php
 
PAW WORKFORCE – March 30-April 1, 2015 (San Francisco)
Speaker proposals due Sept 1 – http://pawcon.com/submit_workforce.php
 
TEXT ANALYTICS WORLD SAN FRANCISCO – March, 2015
Speaker proposals due Sept 1 – http://tawcon.com/call-for-speakers
 
PAW TORONTO – May, 2015
Speaker proposals due Oct 3 – http://pawcon.com/submit.php
 
PAW CHICAGO – June, 2015
Speaker proposals due Jan 9 – http://pawcon.com/submit.php
 
PAW MANUFACTURING – June, 2015 (Chicago)
Speaker proposals due Jan 9 – http://pawcon.com/submit.php

 

No Comments yet »

July 16th 2014

Book Review of ‘Applied Predictive Analytics’ by Dean Abbott

Book Review of 'Applied Predictive Analytics' by Dean Abbott

By: Eric Siegel, Founder, Predictive Analytics World

Industry leader and author Dean Abbott will be presenting at Predictive Analytics World Boston (Oct 5 – 9) on “Data Preparation from the Trenches: 4 Approaches to Deriving Attributes.” Abbott will also run two post-conference full-day training workshops, “Advanced Methods Hands-on: Predictive Modeling Techniques” (where his book Applied Predictive Analytics will be given to attendees), and “Supercharging Prediction: Hands-On with Ensemble Models

Dean Abbott’s new book, “Applied Predictive Analytics: Principles and Techniques for the Professional Data Analyst,” published by Wiley in April, smoothly delivers the established teachings of a preeminent hands-on instructor.

This groundbreaking contribution to the field of predictive analytics provides a unique gift: A how-to that is accessible, yet quite comprehensive, taking the reader through much of the established teachings of one of the industry’s preeminent hands-on instructors. The author, Dean Abbott, is renowned as both a leading “rock star” hands-on consultant in predictive analytics, as well as a fantastic, 5-star-rated conference speaker and an acclaimed training workshop instructor. You get the best of all worlds with this particular expert: deep analytical insights, stellar execution, clear communication, and contagious enthusiasm. And he has translated these assets nicely into a book.

Abbott’s stated mission with this book (as mentioned in its “Introduction” at the end of the book) is to provide very practical guidance for executing on predictive analytics, as if chatting to someone peering over his shoulder as he works through a project. This mission is accomplished, and in doing so it accomplishes something even more significant: The book takes much of Abbott’s well-honed training agenda (do attend his in-person sessions if you can!), along with the accessibility of his casual speaking style, and translates them onto the page. As a result, this book reads in a much more conducive and engaging manner than, say, a more formally structured textbook.

Click here to access the complete article on Predictive Analytics Times.

No Comments yet »

May 29th 2014

PAW Manufacturing Is 3 Weeks Away – Why It Matters

Predictive Analytics World for Manufacturing – June 17-18 in Chicago (www.pawcon.com/mfg), alongside the regular PAW Business event (www.pawcon.com/chicago) – focuses on real-world examples of deployed predictive analytics. Attend and hear how some of the world's largest and most forward-thinking manufacturers are tapping the power of predictive modeling to improve business outcomes. View the agenda-at-a-glance

—————

ARTICLE BY PAW MANUFACTURING CO-CHAIR:

Preparing the Manufacturing Industry for the Upcoming Data Deluge

By Bala Deshpande

The number of computing devices has far exceeded the number of humans on this planet – this milestone was in fact achieved back in 2008. Today there are an estimated 3 processors for every person in the world. Which devices in common use today use the most processors and computing power? The answer – the automobile – may surprise many people. The average car has anywhere between 50-60 microprocessors which handle everything from entertainment systems and climate control to performing advanced engine diagnostics and airbags during crash safety.  

With mobile becoming an integral part of our daily lives, the information processed by these devices becomes useful not only to the manufacturers, but also to consumers. Data from a large and growing number of embedded processors can be used in a variety of ways: at the driver level to select a more personalized, convenient and efficient mode of driving, at a macroscopic (e.g. traffic) level, information collected in this way then can be consolidated to find solutions to problems like traffic jams, helping to improve the flow of traffic should city officials want to leverage this data as well. From a manufacturing perspective, close attention to fine grain data can significantly improve vehicle quality, by potentially reducing warranty and recall issues. The U.S. government has dedicated significant research to this effort and Dr. Thomas Klier of the Federal Reserve of Chicago will be keynoting on this topic at the upcoming PAW-Manufacturing conference. 

This story is repeated across other non-consumer applications, particularly on the shop floor, from small contract manufacturers to the largest original equipment manufacturers of the industry. For example, one small company is using overhead GPS sensors to carefully account for the activities of their assembly line workers. How much time is spent by worker #1 on project #2 which requires inspection versus project #1 which requires assembly? The data generated by the sensors is captured and processed in near real time to predict machine usage, unexpected delays and downtimes and improve overall process efficiency. Once again, the chief output of this connected network is high quality data in huge volumes, which needs to be converted into useful data products. As Surendra Reddy, another keynote speaker from PAW-Mfg says "We are moving away from just creating and accessing data to more intimate, automated, knowledgeable, actionable insights about everything. This allows us to improve machine performance and the efficiency of the systems and networks that link them.  As the devices and environments around us become aware and intelligent, they will play a key role in making decisions for us or nudging us toward specific actions".

Historically manufacturing has been no stranger to either data or analytics. Statistical quality control, operations research and related technologies such as six sigma, all originated in manufacturing and could be considered the predecessors of today's predictive analytics. However much has changed since these technologies became standards. Our ability to collect far more data and in real time has increased significantly. W. David Stephenson, who is another keynote speaker at PAW-Mfg, writes that this upcoming data avalanche from the connected internet of things in a manufacturing plant can transform precision manufacturing by being able to "do things that were impossible in the past – such as identifying and removing defective products while they are still on the conveyor belt. As a result, we had to rely on approximations, such as mean time to failure and scheduled maintenance that inflated costs and often missed outlier conditions that might lead to premature failure".

The goal of PAW-Mfg is to help tie the advances in big data, analytics and visualization to persistent issues to drive orders of magnitude improvements in manufacturing performance and to prepare an "elder statesman" of industry verticals for a connected and data driven future.

Bio:
Bala Deshpande is the co-chair of the inaugural Predictive Analytics World – Manufacturing conference. He is the founder and principal at SimaFore, a custom analytics developer and consulting company. He has nearly two decades of experience in data analytics and related fields. He began his career as an engineering consultant, following which he spent nearly a decade analyzing data from automobile crash tests and helping to build safer cars at Ford Motor Company. He holds a PhD in Bioengineering from Carnegie Mellon and an MBA from Ross School of Business (Michigan). 

No Comments yet »

May 27th 2014

Vote for “Predictive Analytics” Today – Just One Click

There is only one more day to vote (SINGLE-CLICK) for my book.

My book, Predictive Analytics: The Power to Predict Who Will Click, Buy, Lie, or Die, was nominated for a Small Business Trends Book Award!

Show your support in just one click! VOTE HERE (before the end of May 28)

No Comments yet »

March 5th 2014

Breakthrough: How to Avert Analytics’ Most Treacherous Pitfall

This article will make you feel better. And you do need to feel better, if you are one of the many of us who practice analytics—or who must consume and rely on analytics—and find ourselves carrying tension in our shoulders or sometimes losing sleep.

 

The fear stems from a well-known warning of tragic mishap: "If you torture the data long enough, it will confess," as stated by University of Chicago economics professor Ronald Coase. There is a general sense that math could be wrong and that analytics is an art.

 

As John Elder of Elder Research put it, "It's always possible to get lucky (or unlucky). When you mine data and find something, is it real, or chance?" How can we confidently trust what a computer claims to have learned? How do we avert the dire declension, "Lies, damned lies, and statistics"?

 

There is a simple, elegant solution from Elder—but first, let me further magnify your fear: Even the very simplest predictive model risks utter failure. Mistaken, misleading conclusions are in fact terribly easy to come by.

 

A conclusion drawn about one single variable—even without the use of a common multivariate model (such as log-linear regression)—can go awry. In fact, one of the more famous such analytical insights, "an orange used car is least likely to be a lemon," has recently been debunked by Elder and his colleague Ben Bullard at Elder Research, Inc.

 

Big data, with all its pomp and circumstance, can actually mean big risk. More data can present more opportunities to inadvertently discover untrue patterns that appear misleadingly strong within your dataset—but, in fact, do not hold true in general. To be more specific, "bigger" data could mean longer data (a longer list of examples, which generally helps avert spurious conclusions), but also could mean wider data (more columns—more variables/factors per example). So, even if you are only considering one variable at a time, such as the color of each car, you are more likely to come across one that just happens to look predictive in your data by sheer chance alone. This peril that arises when searching across many variables has been dubbed by John Elder vast search.

 

Dr. Elder puts it this way: "Modern predictive analytic algorithms are hypothesis-generating machines, capable of testing millions of 'ideas.' The best result stumbled upon in its vast search has a much greater chance of being spurious…  The problem is so widespread that it is the chief reason for a crisis in experimental science, where most journal results have been discovered to resist replication; that is, to be wrong!"

 

A few years ago, Berkeley Professor David Leinweber made waves with his discovery that the annual closing price of the S&P 500 stock market index could have been predicted from 1983 to 1993 by the rate of butter production in Bangladesh. Bangladesh's butter production mathematically explains 75 percent of the index's variation over that time. Urgent calls were placed to the Credibility Police, since it certainly cannot be believed that Bangladesh's butter is closely tied to the U.S. stock market. If its butter production boomed or went bust in any given year, how could it be reasonable to assume that U.S. stocks would follow suit? This stirred up the greatest fears of PA skeptics, and vindicated nonbelievers. Eyebrows were raised so vigorously, they catapulted Professor Leinweber onto national television.

 

Crackpot or legitimate educator? …

 

Read the full article by Eric Siegel on the Predictive Analytics Times

 

No Comments yet »

Next »

  • Subscribe Via Email

    Get a daily digest of new posts delivered to your inbox:

  • Predictive Analytics Book: The Power to Predict Who Will Click, Buy, Lie, or Die
    BOOK AWARD:
  • Recent Posts