Full Mega-PAW 7-Track Agenda – Detailed Session Descriptions

Predictive Analytics World

June 16-20, 2019 – Caesars Palace, Las Vegas


This page shows the full 7-track agenda for the five co-located conferences at Mega-PAW. Mega Pass registration is required for full access. To view the agenda for one individual conference, click here: PAW Business, PAW Financial, PAW Healthcare, PAW Industry 4.0, or Deep Learning World.

Session Levels:

Blue circle sessions are for All Levels
Red triangle sessions are Expert/Practitioner Level

Day 1 - Tuesday, June 18th, 2019

8:00 am
Registration & Networking Breakfast
Registration & Networking Breakfast
Registration & Networking Breakfast
Registration & Networking Breakfast
Registration & Networking Breakfast
8:45 am
PAW Business
Eric SiegelPredictive Analytics World
Conference Founder
Predictive Analytics World
PAW Financial
Eric SiegelPredictive Analytics World
Conference Founder
Predictive Analytics World
PAW Healthcare
Eric SiegelPredictive Analytics World
Conference Founder
Predictive Analytics World
PAW Industry 4.0
Eric SiegelPredictive Analytics World
Conference Founder
Predictive Analytics World
Deep Learning World
Eric SiegelPredictive Analytics World
Conference Founder
Predictive Analytics World
8:50 am
PAW Business MEGA-PAW SUPER-PLENARY KEYNOTE

A veteran applying deep learning at the likes of Apple, Samsung, Bosch, GE, and Stanford, Mohammad Shokoohi-Yekta kicks off Mega-PAW 2019 by addressing these Big Questions about deep learning and where it's headed:

  • Late-breaking developments applying deep learning in retail, financial services, healthcare, IoT, and autonomous and semi-autonomous vehicles
  • Why time series data is The New Big Data and how deep learning leverages this booming, fundamental source of data
  • What's coming next and whether deep learning is destined to replace traditional machine learning methods and render them outdated
Session description
Speaker
Mohammad Shokoohi-YektaApple
Senior Data Scientist
Apple
PAW Financial MEGA-PAW SUPER-PLENARY KEYNOTE

A veteran applying deep learning at the likes of Apple, Samsung, Bosch, GE, and Stanford, Mohammad Shokoohi-Yekta kicks off Mega-PAW 2019 by addressing these Big Questions about deep learning and where it's headed:

  • Late-breaking developments applying deep learning in retail, financial services, healthcare, IoT, and autonomous and semi-autonomous vehicles
  • Why time series data is The New Big Data and how deep learning leverages this booming, fundamental source of data
  • What's coming next and whether deep learning is destined to replace traditional machine learning methods and render them outdated
Session description
Speaker
Mohammad Shokoohi-YektaApple
Senior Data Scientist
Apple
PAW Healthcare MEGA-PAW SUPER-PLENARY KEYNOTE

A veteran applying deep learning at the likes of Apple, Samsung, Bosch, GE, and Stanford, Mohammad Shokoohi-Yekta kicks off Mega-PAW 2019 by addressing these Big Questions about deep learning and where it's headed:

  • Late-breaking developments applying deep learning in retail, financial services, healthcare, IoT, and autonomous and semi-autonomous vehicles
  • Why time series data is The New Big Data and how deep learning leverages this booming, fundamental source of data
  • What's coming next and whether deep learning is destined to replace traditional machine learning methods and render them outdated
Session description
Speaker
Mohammad Shokoohi-YektaApple
Senior Data Scientist
Apple
PAW Industry 4.0 MEGA-PAW SUPER-PLENARY KEYNOTE

A veteran applying deep learning at the likes of Apple, Samsung, Bosch, GE, and Stanford, Mohammad Shokoohi-Yekta kicks off Mega-PAW 2019 by addressing these Big Questions about deep learning and where it's headed

  • Late-breaking developments applying deep learning in retail, financial services, healthcare, IoT, and autonomous and semi-autonomous vehicles
  • Why time series data is The New Big Data and how deep learning leverages this booming, fundamental source of data
  • What's coming next and whether deep learning is destined to replace traditional machine learning methods and render them outdated
Session description
Speaker
Mohammad Shokoohi-YektaApple
Senior Data Scientist
Apple
Deep Learning World MEGA-PAW SUPER-PLENARY KEYNOTE

A veteran applying deep learning at the likes of Apple, Samsung, Bosch, GE, and Stanford, Mohammad Shokoohi-Yekta kicks off Mega-PAW 2019 by addressing these Big Questions about deep learning and where it's headed:

  • Late-breaking developments applying deep learning in retail, financial services, healthcare, IoT, and autonomous and semi-autonomous vehicles
  • Why time series data is The New Big Data and how deep learning leverages this booming, fundamental source of data
  • What's coming next and whether deep learning is destined to replace traditional machine learning methods and render them outdated
Session description
Speaker
Mohammad Shokoohi-YektaApple
Senior Data Scientist
Apple
9:15 am
PAW Business MEGA-PAW SUPER-PLENARY KEYNOTE

In the United States, between 1500 and 3000 infants and children die due to abuse and neglect each year. Children age 0-3 years are at the greatest risk. The children who survive abuse, neglect and chronic adversity in early childhood often suffer a lifetime of well-documented physical, mental, educational, and social health problems. The cost of child maltreatment to American society is estimated at $124 - 585 billion annually.

A distinctive characteristic of the infants and young children most vulnerable to maltreatment is their lack of visibility to the professionals. Indeed, approximately half of infants and children who die from child maltreatment are not known to child protection agencies before their deaths occur. 

Early detection and intervention may reduce the severity and frequency of outcomes associated with child maltreatment, including death. 

In this talk, Dr. Daley will discuss the work of the nonprofit, Predict-Align-Prevent, which implements geospatial machine learning to predict the location of child maltreatment events, strategic planning to optimize the spatial allocation of prevention resources, and longitudinal measurements of population health and safety metrics to determine the effectiveness of prevention programming.  Her goal is to discover the combination of prevention services, supports, and infrastructure that reliably prevents child abuse and neglect.  

Session description
Speaker
Dyann Daley MDPredict Align Prevent
Founder and CEO
Predict-Align-Prevent
PAW Financial MEGA-PAW SUPER-PLENARY KEYNOTE

In the United States, between 1500 and 3000 infants and children die due to abuse and neglect each year. Children age 0-3 years are at the greatest risk. The children who survive abuse, neglect and chronic adversity in early childhood often suffer a lifetime of well-documented physical, mental, educational, and social health problems. The cost of child maltreatment to American society is estimated at $124 - 585 billion annually.

A distinctive characteristic of the infants and young children most vulnerable to maltreatment is their lack of visibility to the professionals. Indeed, approximately half of infants and children who die from child maltreatment are not known to child protection agencies before their deaths occur. 

Early detection and intervention may reduce the severity and frequency of outcomes associated with child maltreatment, including death. 

In this talk, Dr. Daley will discuss the work of the nonprofit, Predict-Align-Prevent, which implements geospatial machine learning to predict the location of child maltreatment events, strategic planning to optimize the spatial allocation of prevention resources, and longitudinal measurements of population health and safety metrics to determine the effectiveness of prevention programming.  Her goal is to discover the combination of prevention services, supports, and infrastructure that reliably prevents child abuse and neglect.  

Session description
Speaker
Dyann Daley MDPredict Align Prevent
Founder and CEO
Predict-Align-Prevent
PAW Healthcare MEGA-PAW SUPER-PLENARY KEYNOTE

In the United States, between 1500 and 3000 infants and children die due to abuse and neglect each year. Children age 0-3 years are at the greatest risk. The children who survive abuse, neglect and chronic adversity in early childhood often suffer a lifetime of well-documented physical, mental, educational, and social health problems. The cost of child maltreatment to American society is estimated at $124 - 585 billion annually.

A distinctive characteristic of the infants and young children most vulnerable to maltreatment is their lack of visibility to the professionals. Indeed, approximately half of infants and children who die from child maltreatment are not known to child protection agencies before their deaths occur.

Early detection and intervention may reduce the severity and frequency of outcomes associated with child maltreatment, including death.

In this talk, Dr. Daley will discuss the work of the nonprofit, Predict-Align-Prevent, which implements geospatial machine learning to predict the location of child maltreatment events, strategic planning to optimize the spatial allocation of prevention resources, and longitudinal measurements of population health and safety metrics to determine the effectiveness of prevention programming.  Her goal is to discover the combination of prevention services, supports, and infrastructure that reliably prevents child abuse and neglect.  

Session description
Speaker
Dyann Daley MDPredict Align Prevent
Founder and CEO
Predict-Align-Prevent
PAW Industry 4.0 MEGA-PAW SUPER-PLENARY KEYNOTE

In the United States, between 1500 and 3000 infants and children die due to abuse and neglect each year. Children age 0-3 years are at the greatest risk. The children who survive abuse, neglect and chronic adversity in early childhood often suffer a lifetime of well-documented physical, mental, educational, and social health problems. The cost of child maltreatment to American society is estimated at $124 - 585 billion annually.

A distinctive characteristic of the infants and young children most vulnerable to maltreatment is their lack of visibility to the professionals. Indeed, approximately half of infants and children who die from child maltreatment are not known to child protection agencies before their deaths occur.

Early detection and intervention may reduce the severity and frequency of outcomes associated with child maltreatment, including death.

In this talk, Dr. Daley will discuss the work of the nonprofit, Predict-Align-Prevent, which implements geospatial machine learning to predict the location of child maltreatment events, strategic planning to optimize the spatial allocation of prevention resources, and longitudinal measurements of population health and safety metrics to determine the effectiveness of prevention programming.  Her goal is to discover the combination of prevention services, supports, and infrastructure that reliably prevents child abuse and neglect.  

Session description
Speaker
Dyann Daley MDPredict Align Prevent
Founder and CEO
Predict-Align-Prevent
Deep Learning World MEGA-PAW SUPER-PLENARY KEYNOTE

In the United States, between 1500 and 3000 infants and children die due to abuse and neglect each year. Children age 0-3 years are at the greatest risk. The children who survive abuse, neglect and chronic adversity in early childhood often suffer a lifetime of well-documented physical, mental, educational, and social health problems. The cost of child maltreatment to American society is estimated at $124 - 585 billion annually.

A distinctive characteristic of the infants and young children most vulnerable to maltreatment is their lack of visibility to the professionals. Indeed, approximately half of infants and children who die from child maltreatment are not known to child protection agencies before their deaths occur. 
Early detection and intervention may reduce the severity and frequency of outcomes associated with child maltreatment, including death. 
In this talk, Dr. Daley will discuss the work of the nonprofit, Predict-Align-Prevent, which implements geospatial machine learning to predict the location of child maltreatment events, strategic planning to optimize the spatial allocation of prevention resources, and longitudinal measurements of population health and safety metrics to determine the effectiveness of prevention programming.  Her goal is to discover the combination of prevention services, supports, and infrastructure that reliably prevents child abuse and neglect.  

Session description
Speaker
Dyann Daley MDPredict Align Prevent
Founder and CEO
Predict-Align-Prevent
9:40 am
PAW Business Sponsored Session:
The Session Description will be available shortly.
Session description
Sponsored by
dotData
PAW Financial Sponsored Session
The Session Description will be available shortly.
Session description
PAW Healthcare Sponsored Session
The Session Description will be available shortly.
Session description
PAW Industry 4.0 Sponsored Session
The Session Description will be available shortly.
Session description
Deep Learning World Sponsored Session
The Session Description will be available shortly.
Session description
10:00 am
Exhibits & Morning Coffee Break
Exhibits & Morning Coffee Break
Exhibits & Morning Coffee Break
Exhibits & Morning Coffee Break
Exhibits & Morning Coffee Break
10:30 am
PAW Business Track 1: BUSINESS - Analytics operationalization & management
Operationalizing analytics

Multiple surveys show that operationalizing data science, advanced analytics and AI is a major barrier to data-driven decision-making in organizations. Getting even actionable insight across the "last mile" and into operations is hard. In 2018 McKinsey identified that leaders in advanced analytics not only focused on the last mile, they behaved differently. Specifically, they didn't start with the data, but with the decision-making they hoped to change.

In this session that kicks off the Operationalization track, James Taylor analyzes why the last mile is so hard, shares the research that shows how important to success this last mile is, and outlines a practical approach to working backwards to success with data science.

Session description
Speaker
James TaylorDecision Management Solutions
CEO
Decision Management Solutions
PAW Business Track 2: TECH - Machine learning methods & advanced topics
Lifetime value

Customer Lifetime Value (CLV) is considered one of the most useful measures for business to consumer (B2C) companies, and is usually considered more valuable than other measures like conversion rate, average order value, and purchase frequency. If an accurate measure of CLV can be obtained, companies can determine which customers to prioritize with marketing messages and discount offers.

Basic CLV is actually quite easy to compute. But more sophisticated analysts and statisticians use parametric models that take into account purchase frequency, purchase recency, churn risk, and even customer age. These models can provide value estimates 5, 8, and even more than 10 years into the future. However, most retailers, while interested in lifetime value, are especially interested in estimating near-term customer value so they can create effective marketing strategies now.
In this talk, SmarterHQ's founding Chief Data Scientist Dean Abbott describes non-parametric machine learning approaches to calculating customer value for retail that can accommodate additional measurements and features not typically used in CLV models. Model summaries and accuracy metrics for several retail clients will illustrate the effectiveness of this style of model.

Session description
Speaker
Dean AbbottSmarterHQ
Co-Founder and Chief Data Scientist
SmarterHQ
PAW Business Track 3: CASE STUDIES - Cross-industry business applications of machine learning
Airfare prediction
Case Study: Hopper

At Hopper, we predict airfare from a stream of 30 billion daily prices. In this session, we'll talk shop, covering our process for:

  • Personalizing 30 million user conversations through push notifications
  • Measuring user travel flexibility and recommending alternative flights and hotels
  • Building trust with data
  • Open problems
Session description
Speaker
Patrick SurryHopper
Chief Data Scientist
Hopper
PAW Financial
Analytics team building
10:30 am - 10:50 am
Case Study: Manulife

Many companies regardless of their size and years in business may not actually have an analytics team or may have a team of one. During my last speaking engagement at PAW I spoke with a great deal of folks who were interested in creating analytics units but didn't really know how to go about it or were under assumption that it would be very cost prohibitive. This is a case study of a team that consists of former spreadsheet guy, grad with a fresh masters in engineering, a former rocket scientist and two former workflow coordinators.

Session description
Speaker
Richard LeeManulife
Sr Data Scientist
Manulife
Predictive Analytics World for Financial
Credit scoring
10:55 am - 11:15 am
Case Study: PriceWaterhouseCoopers

This project sets forth the work to be performed for the Customer Acquisition Model, aiming to score the entire through-the-door (TTD) populations and eventually make optimal lending decisions. The work will be divided into two phases: Phase 1 will concentrate on building a Minimum Viable Product (MVP). This phase will leverage existing techniques used currently, where applicable, including the target, data sources and transformations. Phase 2 will conduct another round of data exploration with the purpose of identifying additional transformations to increase model lift over what was already achieved in Phase 1.

Session description
Speaker
Chenyu (Jim) GaoPricewaterhouseCoopers
Advisory Manager, Machine Learning/Artificial Intelligence Accelerator
PricewaterhouseCoopers, LLP
PAW Healthcare
Case Study: Zuckerburg San Francisco General Hospital

Bringing the benefits from AI efforts to the frontline workers continues to be a struggle across major healthcare organizations.  We worked on a novel, practical approach to directly take on the workflows of healthcare workers.  This session shares the successes and failures in our attempts, and the AI's introduced via this approach to achieve efficiency and/or outcome goals. This practical workflow approach uses AI as tools, hence can deploy various AI’s for a variety of problems including patient status tracking and task automation. AI's being directly in the workflow also enables continuous learning, process improvement, and optimization toward specific goals.

Session description
Speaker
Cheong AngUniversity of California, San Francisco
Workflow AI Project Lead
University of California, San Francisco
PAW Industry 4.0 KEYNOTE
Case Study: Intel

AI is framed by models, sensors and technologies.  These often ignore the human who must deal with and trust AI outputs.  How do we translate the mental models and senses that humans deploy daily into algorithms that take us from data to inference to action?  With the explosion of sensors at the edge, how do we actually make sense at the edge?  This presentation draws from a recent Intel study of over 250 people in manufacturing and its supporting ecosystem to explore what it takes to accelerate the adoption of Industry 4.0 in a systems of systems approach.

Session description
Speaker
Irene PetrickIntel
Senior Director of Industrial Innovation
Intel Corporation
Deep Learning World KEYNOTE

Over the last few years, convolutional neural networks (CNN) have risen in popularity, especially in the area of computer vision. Many mobile applications running on smartphones and wearable devices would potentially benefit from the new opportunities enabled by deep learning techniques. However, CNNs are by nature computationally and memory intensive, making them challenging to deploy on a mobile device. We explain how to practically bring the power of convolutional neural networks and deep learning to memory and power-constrained devices like smartphones. We’ll illustrate the value of these concepts with real-time demos as well as case studies from Google, Microsoft, Facebook and more. You will walk away with various strategies to circumvent obstacles and build mobile-friendly shallow CNN architectures that significantly reduce memory footprint.

Session description
Speakers
Siddha GanjuNvidia
Solutions Architect
Nvidia
Meher KasamSquare
Software Developer
Square
Anirudh Koulaira
Head of AI & Research
Aira
11:15 am
5-minute transition between sessions
5-minute transition between sessions
5-minute transition between sessions
5-minute transition between sessions
5-minute transition between sessions
11:20 am
PAW Business Track 1: BUSINESS - Analytics operationalization & management
Project management
11:20 am - 11:40 am

Black box algorithms.  Test data with test results. Predictions with possibilities only.  All of these are reminders of analytics teams that have not yet plugged into the “business.”  They appear to be making progress, or at least they are busy, but the results are not tangible.  They have not yet created value that can be measured and replicated.  Although predictive analytics is a “must” for nearly every business today, there are few companies really putting predictive analytics to work for them. 

Why are so many organizations developing predictive capabilities, but haven’t put them to use with their sales, marketing or operations people?  Are companies really getting value from statistical predictions?   If they are, how are they measuring that value and showing it on their bottom line?  If not, what are they doing to close the gap between data science and everyday business.
In this session, we’ll evaluate 3 areas that are most neglected and hardest to deal with when putting predictive analytics to work and show you how to get your predictions used.

  1. Trust.  Getting the users to trust the prediction of an algorithm is fraught with biases.  “That prediction can’t be right because the data is all wrong.”  “I don’t believe that customer will default next month; I am best friends with the CIO and I haven’t heard a word.  Trusting the outcomes of the predictions is the first barrier to overcome. 
  2. Teaching.  Teaching sales and operations people to use the predictions can be your secret to having successful deployments of systems that use your predictions.   Helping users understand the context around the predictions is essential.
  3. Technology.  Automation and self-service are the keys to use of predictive analytics.  It must be easy.  It must produce results.  IT is the it!
Theresa Kushner, partner in Business Data Leadership, comes with over 20 years of experience in deploying predictive analytics at IBM, Cisco, VMware and Dell.  
Session description
Speaker
Theresa KushnerDell EMC
Formerly Sr Vice President, Performance Analytics Group
Dell EMC
Predictive Analytics World for Business
Track 1: BUSINESS - Analytics operationalization & management
Project management
11:45 am - 12:05 pm
Case Study: EMPLOYERS

The success or failure of analytics and data science initiatives often hinges on whether those on the “front lines” of business actually use and follow them.  In this talk the presenter will share ideas he has learned over the years that help maximize the chances of successful analytics deployment. 

Session description
Speaker
Tom WardenEMPLOYERS
SVP, Chief Data and Analytics Officer
EMPLOYERS
PAW Business Track 2: TECH - Machine learning methods & advanced topics
Data preparation

Marketing Predictive Models have seen significant growth in deployments over the past few years with many companies rolling them out for retailers. Marketing data provides many good examples of large robust datasets with clear target variables. It is a common step in model building to do dimensional reduction or variable selection of input fields in order to improve the quality of the models. At SmarterHQ, we have multiple clients with these models in production. Typically these have 100’s of input fields that have overlapping predictive power. To reduce this overlap, many different methods can be deployed including deviation limits, correlation thresholds, stepwise regression, etc. In this talk, we will discuss methods of input variable field selection and its impact on model quality on production data.

Session description
Speaker
William KompSmarterHQ
Senior Data Scientist
SmarterHQ
PAW Business Track 3: CASE STUDIES - Cross-industry business applications of machine learning
Recommedation systems
Case Study: Twitter

Twitter has amazing and unique content that is generated at an enormous velocity internationally. A constant challenge is how to find the relevant content for users so that they can engage in the conversation. Approaches span collaborative filtering and content based recommendation systems for different use cases. This talk gives insight into unique recommendation system challenges at Twitter's scale and what makes this a fun and challenging task.

Session description
Speaker
Ashish BansalTwitter
Senior Engineering Manager
Twitter
PAW Financial
Algorithmic trading
11:20 am -11:40 am
Case Study: Goldman Sachs

Understanding the intraday microstructure dynamics across different universes of stocks and markets is fundamental in designing any optimal  trading strategy, especially for trading diverse portfolios of stocks. In this talk we discuss how modern machine learning techniques can be used in conjunction with dynamical modeling of the intraday phenomena to identify those trading strategies that move away from the general country/sector classifications but rather respect the stock's particular microstructure characteristics.

Session description
Speakers
Andreas PetridesGoldman Sachs
Quantitative Researcher
Goldman Sachs
Michael SteliarosGoldman Sachs
Managing Director
Goldman Sachs
Predictive Analytics World for Financial
Analytics project management
11:45 am - 12:05 pm
Case Study: Safety National Casualty

Understanding the problem to be solved is the most critical element in a successful project. A model that gets 99.3% accuracy to the wrong question does not help the client. And not being able to explain why the results occurred, particularly after a change, does not lead to success. It leads to frustration and the inability to use the model. Safety National's first Data Science Project is a clear example of having the right people in the process at the right time.

Session description
Speaker
William WIlkinsSafety National Casualty Corporation
Chief Risk and Data Analytics Officer
Safety National Casualty Corporation
PAW Healthcare
Case Study: Predicting Cardiovascular Disease

Until recently, healthcare has not understood root causes of diseases well enough for prevention; the main approach has historically been to treat patients after onset. While primary prediction scoring systems are routine for CVD patients, the goal is to reach patients before primary events occur. Amgen and a startup partner are co-developing a machine learning solution that uses existing EMR data to develop statistical and machine learning models predicting secondary CVD events. Having more accurate risk prediction models could significantly impact approaches to disease prevention. Session will also cover role of partnership in sourcing, prototyping, piloting, and scaling novel technologies.

Session description
Speaker
Erich Wohlhieter Ph.DAmgen
Executive Director, Digital Health & Innovation
Amgen
PAW Industry 4.0

Many organizations are faced with the challenge of how to analyze their sensitive data without hosting it on any public cloud. This talk will focus on companies who collect data from their factory operations and are interested in predicting mechanical failures. The audience will get an overview of how to formulate their business problem, perform feature engineering and build a predictive maintenance model using R/Python.

Session description
Speaker
Jaya MathewMicrosoft
Senior Data Scientist
Microsoft
Deep Learning World
Case Study: Trimble, Inc.

Deep learning certainly has roots in the autonomous vehicle space. However, most trucking companies have a substantial investment in existing class 8 semi-trailer trucks that are not going to be replaced overnight. Trimble Transportation Mobility is using deep learning technologies, in conjunction with other advanced analytic techniques and state of the art DevOps approaches to help ensure the safe operation of trucking fleets. While it may be premature for many trucking fleets to embrace autonomous vehicles, TTM has made it possible for those same companies to leverage deep learning as a way to reduce costs and improve safety.

Session description
Speakers
Josh ChapmanTrimble
Data Engineer
Trimble Transportation Mobility
Miles PorterTrimble
Data Scientist
Trimble
Ryan WolbeckTrimble
Data Scientist
Trimble Transportation Mobility
12:05 pm
Lunch
Lunch
Lunch
Lunch
Lunch
1:30 pm
PAW Business KEYNOTE

How many .edu addresses are in your inbox right now? As organizations pursue digital transformation strategies, challenges related to finding and retaining analytical talent, objectively assessing the relevance of new, and emerging technology and engaging in deep and meaningful innovation with eventual payback are common to all sectors of the economy. Deep, collaborative partnerships with universities can help mitigate many of these challenges. This is all the more true because data science itself has given rise to a new "entrepreneurial university" paradigm. Dr. Priestley is an academic Associate Dean, who worked for organizations like Accenture and VISA EU, and now manages corporate partnerships with the likes of Blue Cross Blue Shield, Emerson, Equifax, and GE, as well as fire departments and law enforcement. She will discuss the ways that organizations should be thinking about working with universities, but typically don’t – including research, innovation, "externships," training options, recruitment, and other strategic relationships. After this session, you will never look at universities the same way again.

Session description
Speaker
Jennifer Lewis PriestleyKennesaw State University
Professor of Applied Statistics and Data Science
Kennesaw State University
PAW Financial KEYNOTE
Case Study: Wells Fargo

Financial institutions have long excelled at "analysis." We are overrun with reports, dashboards, key performance indicators and other metrics, and many financial organizations have a history of using this information to make data-driven decisions. But it is no longer enough. With rapid advances in technology and ever increasing stores of data, the opportunity is present for data scientists to dig in and get their hands dirty building real products and services for actual customers. It is time to trade information for engineering. This talk explores the shift from a consulting, insights-generation mindset for analytics to one of data-driven software development and what that means for financial institutions. How can you best structure these new hybrid analytics-engineering teams and how should you set them loose to generate value for the organization? Come to hear more.

Session description
Speaker
Nathan SusanjWells Fargo
Vice President, Data Science Manager
Wells Fargo
PAW Healthcare KEYNOTE

With major new players (Amazon, JP Morgan, Berkshire Hathaway), reconfigured players (CVS merged with Aetna), and lots of hospital consolidation, healthcare is going to change. We are on the cusp of a post-hospital era where advanced analytics will enable and support pay for performance, value-based purchasing, pricing optimization, wellness/disease management, evidence-based medicine, and workforce optimization. Meanwhile the government retools its metrics every few years and tries to keep up.  

This keynote will confront the role of health analytics as a major force in the changing health care landscape.  Professor Rossiter will explain how we are finally entering the post-hospital era, and how all of this will enable the long-awaited managed competition approach to health services delivery.

Session description
Speaker
Louis F Rossiter Ph.DWilliam & Mary School of Business
Professor of Public Policy at the College of William & Mary, former Secretary of Health & Human Resources for the Commonwealth of Virginia, former Deputy for Policy to the Administrator of the Center
College of William & Mary
PAW Industry 4.0

Field issue (malfunction) incidents are costly for the manufacturer’s service department. A normal telematics system has difficulty in capturing useful information even with pre-set triggers. In this session, Yong Sun will discuss how a machine learning, deep learning based predictive software/hardware system has been implemented to solve these challenges by 1) identifying when a fault will happen 2) diagnosing the root cause on the spot based on time series data analysis. Yong Sun will cover a novel technique for addressing a lack of training data for the neural network based root cause analysis.

Session description
Speaker
Yong SunIsuzu
Supervisor
Isuzu Technical Center of America
Deep Learning World
Case Study: Isuzu Technical Center of America

Field issue (malfunction) incidents are costly for the manufacturer’s service department. A normal telematics system has difficulty in capturing useful information even with pre-set triggers. In this session, Yong Sun will discuss how a machine learning, deep learning based predictive software/hardware system has been implemented to solve these challenges by 1) identifying when a fault will happen 2) diagnosing the root cause on the spot based on time series data analysis. Yong Sun will cover a novel technique for addressing a lack of training data for the neural network based root cause analysis.

Session description
Speaker
Yong SunIsuzu
Supervisor
Isuzu Technical Center of America
2:15 pm
PAW Business Sponsored Session
The Session Description will be available shortly.
Session description
Sponsored by
DataRobot
PAW Financial Sponsored Session

Come and learn how predictive analytics can be combined to Decision Management and Notation (DMN) standard to provide business end users with a simple and clear depiction of their business decision context. Using a loan pre-qualification example we will demonstrate how DMN can leverage predictive analytics models captured in PMML. We will also show how the audit data generated from the DMN execution can be used for dashboarding, business intelligence, or fed back into the predictive model. This approach will enable you to clearly demonstrate to business users how predictive models directly contributes value into the day to day business operational decisions.

Session description
Sponsored by
Trisotech
Speaker
Simon RinguetteTrisotech
Senior Architect
Trisotech
PAW Healthcare Sponsored Session
The Session Description will be available shortly.
Session description
Sponsored by
diwo
PAW Industry 4.0 Sponsored Session
The Session Description will be available shortly.
Session description
Sponsored by
diwo
Deep Learning World Gold Sponsored Session
The Session Description will be available shortly.
Session description
2:35 pm
5-minute transition between sessions
5-minute transition between sessions
5-minute transition between sessions
5-minute transition between sessions
5-minute transition between sessions
2:40 pm
PAW Business Track 1: BUSINESS - Analytics operationalization & management
Sourcing analytics staff

In a world where demand outpaces supply, finding and keeping analytics talent has become a real dilemma.  Identifying the right mix of business skills and analytics skills can feel like an impossible search.  With so many people looking for strong talent, it often becomes difficult to compete.  How do you attract the right skills to your team to ensure a strong analytics capability?  What types of levels, roles, and titles do you need?  What are some of the ways to ensure you retain your analytics talent?  This session will discuss different compositions of successful analytics teams, as well as titles, career paths, and tips to win at the salary game. 

Session description
Speaker
Anne G. RobinsonKinaxis
Chief Strategy Officer
Kinaxis
PAW Business Track 2: TECH - Machine learning methods & advanced topics
Active learning

Today, with always more data at their fingertips, Machine Learning experts seem to have no shortage of opportunities to create always better models. Over and over again, research has proven that both the volume and quality of the training data is what differentiates good models from the highest performing ones.

But with an ever-increasing volume of data, and with the constant rise of data-greedy algorithms such as Deep Neural Networks, it is becoming challenging for data scientists to get the volume of labels they need at the speed they need, regardless of their budgetary and time constraints. To address this “Big Data labeling crisis”, most data labeling companies offer solutions based on semi-automation, where a machine learning algorithm predicts labels before this labeled data is sent to an annotator so that he/she can review the results and validate their accuracy.

Unfortunately, even this approach is not always realistic to implement, for example in the context of some industries, such as Healthcare, where obtaining even a single label can cost thousands of dollars.

There is a radically different approach to this problem which focuses on labeling “smarter” rather than labeling faster. Instead of labeling all of the data, it is usually possible to reach the same model accuracy by labeling just a fraction of the data, as long as the most informational rows are labeled. Active Learning allows data scientists to train their models and to build and label training sets simultaneously in order to guarantee the best results with the minimum number of labels. In this talk, I will cover both the promises and challenges of Active Learning, and explain why, all in one, Active Learning is a very promising approach to many industry problems.

Session description
Speaker
Jennifer PrendkiAtlassian
VP of Machine Learning
Figure Eight
PAW Business Track 3: CASE STUDIES - Cross-industry business applications of machine learning
Sales analytics; deployment
Case Study: Southwest Airlines

Internally at SWA, predictive scores are delivered to the sales team via documents known as “Quick Start Guides.” The point of these guides is to take an analytics example and repeat it across different hypotheses about the business, 40 of them. 

One example of this is a set of models we built to predict the trajectory of YoY growth for individual accounts to see if they will continue with the same YoY growth or go another direction. 

While that information on it's own is a cool prediction it doesn't service the 'boots on the ground', so we built guides that help the users understand why an account has come to their attention and talking points for those influential attributes so the sales force can use them in client conversations.

Right now the projected gain is $15MM in incremental future revenue per year - just by focusing on educating the frontline sales force.

Session description
Speakers
Bill HoffmanElicit
Chief Growth Officer & Employee Practice Leader
Elicit, LLC
Padrum PanbechiSouthwest Airlines
Sales Analytics Lead
Southwest Airlines
Nick PylypiwElicit
Data Science Manager
Elicit, LLC
PAW Financial
Compliance risk management
Case Study: Bank of America Merrill Lynch

One of the biggest challenges faced by analysts or strategists who are a part of the investment research team in a stock brokerage or investment bank is the accuracy of the equity research reports published by the firm. This session will cover the importance of a system based on natural language processing, and how it was developed in the context of processing pre-released reports and flagging entities and language of interest in an automated manner that fit the framework of a Supervisory Analyst's workflow including legal considerations. We will also cover the compliance review process conducted in parallel.

Session description
Speaker
Bhakthi LiyangeBank of America Merill Lynch
VP, Data Science Lead
Bank of America Merill Lynch
PAW Healthcare
Case Study: Breaking Down the Models

A frequent criticism of the use of machine learning models as compared to human analysis is that ML models are "black boxes" and uninterpretable. Recent advancements in the field of explainable AI allow us to understand what factors influenced both individual predictions and aggregate model behaviors. We will revisit a case study from another PAW conference on predicting hospital readmissions, except this time we will use open-source software and dive into the 'why' with various visualizations that explain the model's behavior.

Session description
Speaker
Cal ZemelmanCVP
Director, Lead Data Scientist
CVP
PAW Industry 4.0
Case study: Autoloop, LLC

We will discuss the approach used to learn topics and their place in a multi level hierarchy on hundreds of millions of text records. These methods are generalizable beyond the domain in which they were applied. We used a combination of supervised and unsupervised machine learning methods, which we will discuss at more length including the technologies, algorithms, and results.

Session description
Speaker
Wes MadrigalAutoLoop
Machine Learning Engineer
AutoLoop
Deep Learning World
Case Study: Adobe

Machine learning has been sweeping our industry, and the creativity it is already enabling is incredible. On the flip side there has also been the emergence of technology like Deep Fakes with the possibility to spread disinformation. As a tool maker, is our technology neutral, or are we responsible for creating technology for good? How should we be thinking about biases of multiple forms when training AI? What can go wrong when learning is applied to indiscriminate user data? 

 At Adobe we look at this problem from multiple angles, from weighing the positives of technology against their possible misuses, researching detection technology for manipulated images, assembling diverse teams of experts, and having internal training and reviews of technology around Artificial Intelligence.

Session description
Speaker
Steve HoegAdobe Systems
Director of Engineering
Adobe
3:25 pm
Exhibits & Afternoon Break
Exhibits & Afternoon Break
Exhibits & Afternoon Break
Exhibits & Afternoon Break
Exhibits & Afternoon Break
3:55 pm
PAW Business Track 1: BUSINESS - Analytics operationalization & management
Strategic analytics
3:55 pm - 4:15 pm
Case Study: Humana Military

In the government contracting world, executives default to using domain knowledge to answer strategic questions. Industry experts are skeptical about using predictive analytics.  But to remain an industry leader, Humana Military needs a broader perspective to diversify and grow its business.

In this talk, hear about how one executive's curiosity about analytics led to a great partnership between executives and data scientists.  Through predictive analytics, we discovered new federal opportunities, uncovered what it takes to win contracts, scored our chances of win, and discovered potential partnerships.  The result: expanding our view of what is possible and co-creating our future together.

Session description
Speaker
Lily Quinto-BantonHumana
Lead Data Scientist
Humana Military
Predictive Analytics World for Business
Track 1: BUSINESS - Analytics operationalization & management
Cross-enterprise adoption
4:20 pm - 4:40 pm
Case Study: Caesers

Caesars Entertainment is the world's most geographically diversified casino-entertainment company with major revenue streams from  restaurants, entertainment, and hotels in addition to gaming. Caesars' VP of Gaming, Data Science and Fraud Analytics will cover some of the predictive analytics questions Caesars faces and approaches used to address these questions. Topics covered include how Caesars is using deep learning to interpret visual data, predicting key marketing characteristics including future spending and profitability, using machine learning for fraud detection, applying predictive analytics to sportsbook decisions, and valuing entertainment's impact on other parts of the business.

Session description
Speaker
Josh FrankCaesars Entertainment
Vice President-Gaming Data Science & Fraud Analytics
Caesars Entertainment
PAW Business Track 2: TECH - Machine learning methods & advanced topics
Workforce analytics

Career rewards -- the long-term value of employment reflecting the trajectories of advancement and pay -- can be used strategically to motivate performance and improve retention. Too often, they are neglected by reward practitioners who focus on benchmarking, and are not the result of deliberate design. In this session, we'll use case studies to show how to measure both the strength and impact of career rewards to optimize the career component of total rewards. In addition, we will demonstrate a methodology that quantifies organization shape in a way that permits alignment with pay on an empirical basis. The session will demonstrate how advanced analytics can inform rewards strategy.

Session description
Speakers
Haig NalbantianMercer
Senior Partner, Co-leader Mercer Workforce Sciences Institute
Mercer
Tauseef RahmanMercer
Principal, Workforce Strategy & Analytics
Mercer
PAW Business Track 3: CASE STUDIES - Cross-industry business applications of machine learning
Marketing analytics
Case Study: Overstock.com

At Overstock.com, lack of data has never been an issue. We know everything from the color you search most, to which room you'll redesign next. We can see individuals transition from furnishing their first flat to building their dream home, but processing this data requires some serious firepower. It has fueled our focus on delivering real-time personalization through the unification of data and AI. 

Tune in as Chris Robison and Mike Finger takes you through martech innovations in building a successful marketing technology infrastructure for instantaneous individualized marketing experiences.

Session description
Speakers
Michael FingerOverstock
Software Architect, Marketing
Overstock.com
Chris RobisonOverstock
Director of Data and Audience
Overstock.com
PAW Financial
Machine learning methods

In this session, I will provide an overview of logistic regression, GLM logistic regression, decision tree, random forest, gradient boosting, neural networks, etc. Then, a comparison will be made through a case study about building a full life cycle of predictive model based on insurance datasets. Since the business goal through the feature engineering stages are similar given the same case study, the comparison of each method, its  advantages and disadvantages, will include the feature selection, model building, model validation, and model testing stages. Also, model implementation and interpretability will be discussed and compared. Finally, we will discuss implementation of these methods in Python and R.

Session description
Speaker
Mei Najim
CSPA, Founder and Lead Data Scientist
Advanced Analytics Consulting Services, LLC
PAW Healthcare

Emergency departments have seen a dramatic increase in the number of visits from elderly patients. Many elderly use a personal emergency response system (PERS) to signal for help in case of an incident such as a fall or breathing problems. At Partners Healthcare, we are testing a predictive model that uses PERS data to predict elderly at high risk of emergency department visits. Clinical staff from our homecare program perform interventions with high-risk patients. This presentation will cover the development of the predictive model and its deployment in a randomized controlled trial.

Session description
Speakers
Sara Golas
Senior Data Specialist
Partners HealthCare Pivot Labs
Jorn op den BuijsPhilips Research
Senior Scientist
Philips Research
PAW Industry 4.0
Case study: Eaton
3:55 pm - 4:15 pm

Industry 4.0 can suffer from a real-world application problem when industrial and manufacturing companies only view IoT as a solution during large-scale plant upgrades or new construction. This case study presents how a manufacturing company has been able to generate energy cost savings in the scale of $MM through targeted deployment of sensors and IoT connected equipment into existing, large, (and sometimes very old and dirty) machinery and factories. By lowering the threshold of what projects are considered worthy of an IoT investment, what was previously considered run-of-the-mill operations can suddenly provide insightful information and an exciting ROI.

Session description
Speakers
Steve MillerEaton
Vice President, Controls and Protection Division
Eaton
Lyle SprinkleEaton
Director: Meters, Relays & IoT Solutions
Eaton
Predictive Analytics World for Industry 4.0
KEYNOTE
4:20 pm - 4:40 pm

IOT is untappable. But there is nothing the hive mind can't do. Particularly with a good algorithm.

Session description
Speaker
Steven RamirezBeyond the Arc
CEO
Beyond the Arc
Deep Learning World
Case Study: Bank of America Merill Lynch

At Bank of America Merrill Lynch, hundreds of equity reports are being produced by hundreds of research analysts in a given day focusing on a specific stock, industry sector, a currency, commodity or fixed-income instrument, or even on a geographic region or country. An exhaustive pre-release review process is in-place that guarantees the factual accuracy of the published report and other regulatory and compliance requirements. The review work, conducted by Supervisory Groups, is largely manual and has to balance the workload requirements of a comprehensive and detailed scrutiny with ever increasing pressure to reduce time to market.

This session will focus on demonstrating how an intelligent document classification system was developed using NLP and how the system leveraged doc2vec and word2vec for creating distributed semantic spaces that provides the context based insights of the documents. This session also will demonstrate how semantics were used to develop deep learning network for sentence classification that flags and identifies questionable entities and language of interest of documents in an automated manner.

Session description
Speaker
Bhakthi LiyangeBank of America Merill Lynch
VP, Data Science Lead
Bank of America Merill Lynch
4:40 pm
5-minute transition between sessions
5-minute transition between sessions
5-minute transition between sessions
5-minute transition between sessions
5-minute transition between sessions
4:45 pm
PAW Business Track 1: BUSINESS - Analytics operationalization & management
Cross-enterprise adoption
4:45 pm - 5:05 pm
Case Study: Pacific Life

Pacific Life has made great strides recently in adoption of analytics across the enterprise. This talk will discuss how the organization took talented and separate analytics practices, built a unified vision, accelerated insights and enhanced adoption at all levels. Specific take-aways for the audience will be around driving stakeholder buy-in, building consensus of vision, getting demonstrable value, and tracking iterative wins. Specific frameworks, anecdotes and examples will be used to engage the audience and create actionable best practices.

Session description
Speaker
Robert M. HorrobinPacific Life
AVP of Advanced Analytics and Planning, Retirement Solutions Division
Pacific Life
Predictive Analytics World for Business
Track 1: BUSINESS - Analytics operationalization & management
Project management
5:10 pm - 5:30 pm

In the age of machine learning, when business stakeholders demand both high accuracy and transparency in predictive models, practitioners must adapt in terms of how they present findings. Evaluation must be applied at all stages in the machine learning workflow -- from the initial POC through the model deployed in production. Each stage places different demands on the metrics we choose, as well as how we communicate and interpret those metrics. This talk will explore this issue and help both developers and product managers navigate the machine learning evaluation landscape.

Session description
Speaker
Leslie BarrettBloomberg
Senior Software Engineer
Bloomberg LP
PAW Business Track 2: TECH - Machine learning methods & advanced topics
Social science analytics
4:45 pm - 5:05 pm

We are living at the dawn of big social science. Just like physics has the particle accelerators and astronomy has orbital telescopes, social scientists can now harness big data and machine learning systems of immense complexity and cost to measure and predict what society is up. Dstillery had been building one such system for the last decade. In this session, I'll walk you through how the system evolved from it's roots in programmatic advertising, how we discovered we were at war with fraudulent data, and how we settled on our philosophy that making good decisions on hundreds of billions of individual pieces of data yields the best results, but at the cost of significant infrastructure and system complexity. Finally, we'll talk about how these shiny new systems don't replace traditional social science methodologies such as surveys, but instead supplement and reinforce them.

Session description
Speaker
Peter LenzDstillery
Senior Geospatial Analyst
Dstillery
Predictive Analytics World for Business
Track 2: TECH - Machine learning methods & advanced topics
Best Practices
5:10 pm - 5:30 pm

Preeminent consultant, author and instructor Dean Abbott, along with Rexer Analytics president Karl Rexer, field questions from an audience of predictive analytics practitioners about their work, best practices, and other tips and pointers.

Session description
Speakers
Dean AbbottSmarterHQ
Co-Founder and Chief Data Scientist
SmarterHQ
Karl RexerRexer Analytics
President
Rexer Analytics
PAW Business Track 3: CASE STUDIES - Cross-industry business applications of machine learning
Marketing analytics
4:45 pm - 5:05 pm

Attribution is about crediting touchpoints in customer interactions with their impact in the sale process, hence the core element of performance marketing. But today, the choice of the model is often driven by subjective belief and guessing, rather than data and analytics. This explains why to date we often find in place relatively basic models, like last-click or last-non-direct. In this session, we will discuss the different models seen in practice, analyze how they perform in different contexts, explore are their core ideas (from statistics, game theory, marketing science and machine learning), and cover their pros & cons. Finally, we will discuss how to turn descriptive attribution into successful predictive analytics.

Session description
Speaker
Alwin Haensel, Ph.D.Haensel AMS
CEO & Founder
Haensel AMS
Predictive Analytics World for Business
Track 3: CASE STUDIES - Cross-industry business applications of machine learning
Cross-industry case studies
5:10 pm - 5:30 pm

There’s a new sense of urgency from the C-Suite to capture more value from the company’s data. For many organizations, this means accelerating progress toward machine learning. But what does it take to go faster? And can you skip some of the steps in an otherwise steep learning curve? 

Drawing on case studies in banking and utilities, this session will provide insights to:

  • Recognize where a project fits in the data science lifecycle
  • Avoid predictive analytics projects that waste time and money  
  • Create an action plan that helps you reap the benefits of machine learning
Session description
Speaker
Steven RamirezBeyond the Arc
CEO
Beyond the Arc
PAW Financial
Advanced methods; forecasting; credit scoring
Case Study: Pacific Life

Wavelets have long been known for their strong ability in denoising and transforming signals, for example in speech recognition and image processing. More recently, Long Short Term Memory Networks (LSTMs) and Autoencoders started showing promising results in this space as well. Application is possible also to data sets generated by complex nonlinear processes with a low signal to noise ratio. Both Wavelets and LSTMs offer value through generating cleaner training data and ultimately driving deeper insights. In this session, we will show three concrete use cases: Analysis of Financial Time Series, Sales Forecasting and Credit Risk Analysis.

Session description
Speaker
Olaf MenzerPacific Life
Senior Data Scientist, Decision Analytics
Pacific Life
PAW Healthcare
Case Study: AI for Anomaly Detection

An emerging biotech company launched the first treatment option in an unestablished rare disease market. Extremely low prevalence, lack of physician awareness, no codified ICD-10 diagnosis code, and the lack of approved treatments resulted in significant mis-diagnosis, making the application of AI challenging. Addressing the challenge required combining first and third party de-identified data in a HIPAA-compliant workflow based on Swoop's prIvacy platform. Of the 84 start forms in the past 6 months, 24 (29%) were due to Swoop's AI model. Further validation is underway in a university hospital system, embedding predictions into clinical workflows to improve patient outcomes.

Session description
Speakers
Dan FisherSwoop
VP Life Sciences
Swoop
Steve LambIPM.ai
Principal
IPM.ai
PAW Industry 4.0
Case study: Optoro
4:45 pm - 5:05 pm

Optoro’s three core data culture problems were the following:

• Fear of Data
• Inconsistent Use of Vocabulary and Metrics
• Data Mistrust

This presentation will outline the strategies that we used to combat these issues. This on-going endeavor is yielding benefits to Optoro, including
• Increased alignment on company goals
• Improved ease of communication between teams and with Senior Management
• Consistency across external messaging

Session description
Speaker
Vanessa LamOptoro
Manager, Business Intelligence
Optoro
Predictive Analytics World for Industry 4.0
5:10 pm - 5:30 pm

The session helps participants understand the role of data, the importance of a data strategy in an organization, the types of business analytics to execute, and ten practical steps to develop a data strategy to improve business processes. We will explore how to conduct research to identify opportunities to minimize threats, manage risks, and improve performance. 

The session will use the dataFonomics® (A Data to Information Economics Framework) methodology as a platform for developing a Data Strategy. Attendees will be given an overview of the framework, steps, practical knowledge on data analysis and how to develop a data strategy.

Session description
Speaker
Nicole Antoinette SmithOhio University
Lecturer
Ohio University
Deep Learning World
4:45 pm - 5:05 pm
Case Study: Reinsurance Group of America (RGA)

How much data is enough to build an accurate deep learning model? This one of the first and most difficult questions to answer early in any machine learning project. However, the quality and applicability of your data are more important considerations than quantity alone. This talk presents some insights and lessons learned for gauging the suitability of electronic health record (EHR) training data for a life underwriting project. You will see how to determine if more data might increase accuracy and how to identify any weaknesses a deep neural network might have as a result of your current training data.

Session description
Speaker
Jeff HeatonRGA
VP, Data Scientist
Reinsurance Group of America
Deep Learning World
5:10 pm - 5:30 pm

There have been ten US recessions since 1950. When is the next one? The answer matters because asset prices plunge in recessions, which creates both risk and opportunity. Forecasters answer this question by looking at leading economic indicators. We translate the thinking of forecasters into machine learning solutions. This talk explains the use of recurrent neural networks, which excel at learning historical patterns that don’t repeat, but rhyme. Our model anticipates the Great Recession from past data and exhibits lower error than established benchmarks. The proposed approach is broadly applicable to other prediction problems such as revenue and P&L forecasting.

Session description
Speaker
Arnab ChakrabartiHitachi America
Senior Research Scientist
Hitachi America, Ltd.
5:30 pm
Networking Reception
Networking Reception
Networking Reception
Networking Reception
Networking Reception
7:00 pm
End of first Conference Day
End of first Conference Day
End of first Conference Day
End of first Conference Day
End of first Conference Day
Day 2 - Wednesday, June 19th, 2019

8:00 am
Registration & Networking Breakfast
Registration & Networking Breakfast
Registration & Networking Breakfast
Registration & Networking Breakfast
Registration & Networking Breakfast
8:45 am
PAW Business

Comedy from Yoram Bauman, PhD, “the world’s first and only stand-up economist”.

Session description
Yoram Bauman
The world's first and only stand-up economist
PAW Financial
Mei Najim
CSPA, Founder and Lead Data Scientist
Advanced Analytics Consulting Services, LLC
PAW Healthcare
Jeff DealElder Research
Chief Operating Officer
Elder Research
 
8:55 am
PAW Business KEYNOTE

We’re in a global analytics arms race, where yesterday’s strategic advantage can quickly become tomorrow’s industry standard.  To stay competitive, companies must continue to invest and evolve at an ever increasing rate.

In this keynote session, Disney Sr. Vice President of Revenue Management and Analytics, Mark Shafer, will discuss his 30-year rags to riches analytical journey, including lessons learned from being on the receiving end of analytics at People Express Airlines to building a science-based analytical team at The Walt Disney Company.

During his 23 years at Disney, Mark led an analytical transformation, starting by implementing Walt Disney World's first resort revenue management model to currently leading an Internal consulting team of more than 150+ employees responsible for supporting analytics across The Walt Disney Company, including Parks and Resorts, Media Networks (ABC, ESPN, Disney Channel, A&E Networks etc.), Studio Entertainment (The Walt Disney Studios, Disney Theatrical).

Leave with deep insights and practical advice on how to steer a successful analytics journey at your company.

Session description
Speaker
Mark ShaferWalt Disney Company
Senior Vice President Revenue Management and Analytics
Walt Disney Company
PAW Financial KEYNOTE
Claims processing
Case Study: ESIS

Predictive modeling continues to play an important role in the claim process for Property & Casualty (P&C) insurers and Third Party Administrators (TPAs).  This session focuses on the TPA environment utilizing the workers' compensation line of business as a case study into the rationale, implementation, and outcomes that follow the decision to deploy predictive modeling. The speaker will explain the environment of workers' compensation claims handling and the role of the TPA as it relates to assisting employers in managing risk.  With this foundation, the session will move into how predictive modeling can be deployed in the claim process and specifically look at one TPAs efforts to increase efficiency and improve client outcomes.  Participants of the session will walk away with the following information:

  • Basics on the P&C TPA environment
  • Opportunities for predictive modeling in the claim process
  • Implementation challenges and pitfalls to consider regardless of industry
  • Overcoming pitfalls
  • Realizing outcomes in a dynamic environment
Session description
Speaker
Keith HigdonESIS
President
ESIS
PAW Healthcare

Data underlies all of our best efforts to evolve health care practices. Data, and lots of it, now come in many forms and from many sources. Data is the catalyst for the transition from volume-based, episodic care to value-based, personalized care. A workable data strategy has to account for a variety of data forms and sources. A good data strategy bakes in empathy for each individual represented by the data. And, a great data strategy ensures that any movement of data within the organization is reliable, timely, and makes provision for increased data asset value.  Great data strategy is the foundation for improving the delivery and outcomes of our healthcare experience. Gerhard Pilcher will share insights, tips, and lessons learned from more than 20 years of work solving problems and providing guidance to many different types of complex organizations within the health care industry and beyond.

Session description
Speaker
Gerhard PilcherElder Research
President & CEO
Elder Research
PAW Industry 4.0 KEYNOTE
Case study: UPS

Turning data into a business advantage through optimization is the goal of most organizations. UPS has been on a twenty year journey to achieve this goal and has seen cost improvements reaching $1B annually. At the same time, UPS has been able to offer new products and services backed by data and analytics. 

Session description
Speaker
Jack LevisUPS
Senior Director, Process Management
UPS
Deep Learning World KEYNOTE

Deep neural networks provide state-of-the-art results in almost all image classification and retrieval tasks. This session will focus on the latest research on active learning and similarity search for deep neural networks and how they are applied in practice by the Verizon Media Group. Using active learning, we can select better images and substantially reduce the number of images required to train a model. It enables us to achieve state-of the art performance while substantially reducing cost and labor. By using triplet loss for similarity search, we can improve our ability to retrieve better images for shopping application and advertising.

Session description
Speaker
Armin KappelerVerizon
Sr. Research Engineer
Verizon Media Group
9:40 am
PAW Business Sponsored Session
The Session Description will be available shortly.
Session description
Sponsored by
diwo
PAW Financial Sponsored Session
The Session Description will be available shortly.
Session description
Sponsored by
diwo
PAW Healthcare Sponsored Session
The Session Description will be available shortly.
Session description
Sponsored by
DataRobot
PAW Industry 4.0 Sponsored Session
The Session Description will be available shortly.
Session description
Deep Learning World Sponsored Session
The Session Description will be available shortly.
Session description
10:05 am
PAW Business Track 1: BUSINESS - Analytics operationalization & management
Ethics
10:05 am - 10:25 am

Concerns are constantly being raised about what data is appropriate to collect and how (or if) it should be analyzed. There are many ethical, privacy, and legal issues to consider and no clear standards exist in many cases as to is fair and what is foul. This means that organizations must consider their own principles and risk tolerance in order to implement the right policies. This talk will cover a range of ethical, privacy, and legal issues that surround analytics today. It will frame big questions to consider while providing some of the tradeoffs and ambiguities that must be addressed.

Session description
Speaker
Bill FranksInternational Institute For Analytics
Chief Analytics Officer
International Institute For Analytics
Predictive Analytics World for Business
Track 1: BUSINESS - Analytics operationalization & management
Legal ramifications of HR analytics
10:30 am - 10:50 am
Case Study: Jackson Lewis (law firm)

From predicting which candidates will make great employees and which employees are likely to leave the organization, to forecasting diversity trends and achieving pay equity, employers are increasingly turning to data science to help streamline their employment processes. Despite great promise, using data science in workplace management can expose employers to a crippling degree of legal risk and potential liability, if the relevant legal and ethical issues are not carefully considered. Join us for this engaging workshop as a data scientist and a lawyer from preeminent workplace law firm Jackson Lewis demonstrate how employers can unlock the full potential of leveraging data science to manage the workplace and avoid the unintended consequences of doing so.

Session description
Speaker
Eric J. FelsbergJackson Lewis P.C.
Principal, Long Island Office
Jackson Lewis P.C.
PAW Business Track 2: DATA - Data Strategies & data prep
Data strategies
10:05 am - 10:25 am
Case Study: Nike, Infinity

As the global volume of data increases, the challenge of monetizing data is only growing. In fact, data is projected to increase ten-fold by 2025, and 25% will be real-time in nature, requiring sophisticated systems and processes to capture and utilize effectively.   One of the most common business questions overheard at companies is how to leverage the value of “dead data.”  Data monetization is “the collection and packaging of data (or data insights) for delivering value-added services or creating revenue-generating products”. As the term “value” suggests, data monetization goes beyond just selling or transferring data assets. Instead, the best data monetization practices include both direct strategies and indirect strategies. An indirect strategy may involve using data to improve customer experience, drive cross-selling, or improve performance, and a direct strategy may involve creating new sources of revenue with outside partners.

As the volume of data explodes, companies are finding creative ways to exploit this information. During this discussion, Lawrence will talk through simple steps to start leveraging the value of your data, with a specific focus on analytics initiatives.

Session description
Speaker
Lawrence CowanCicero Group
Senior Partner & COO
Cicero Group
Predictive Analytics World for Business
Track 2: DATA - Data Strategies & data prep
Third-party data selection; email marketing analytics
10:30 am - 10:50 am
Case Study: A large insurance company​

This presentation provides insights on how to optimize marketing campaigns by predicting the responses. The original idea was implemented for a large insurance company’s marketing campaign. We modified and perfected the idea, iterated and perfected it for the internal marketing lead generation campaigns. In this case, we gained access to customers’ attributes from a 3rd party data provider and how responders’ responded to previous marketing campaigns. The attributes include: customer age, professions, preferred contact types, months, past campaigns, etc., the target variable was if the customer responded to a previous campaign and purchased an item. We developed multiple machine learning models such as Ensemble, Gradient Boost, etc. We selected the best model with the highest accuracy and finally created the appropriate label for the response. The process allowed us to gain access to a precise marketing list for the campaigns, improving the performance response 25%-30% from the previous campaigns.  

Session description
Speaker
Dan Sarkar
Director of Data Science
Insight Que Solutions
PAW Business Track 3: CASE STUDIES - Cross-industry business applications of machine learning
Political campaign analytics

In this revealing session, Daniel Porter -- BlueLabs co-founder and former Obama for America director of statistical modeling -- will discuss how the current hyperpartisan environment means that the success of your message is strongly determined by how it fits into partisan narratives, increasing the importance of 1:1 message targeting.

Session description
Speaker
Daniel Porter
Co-Founder
BlueLabs
PAW Financial
Model interpretability
Case Study: Paychex

The use of AI in decision making processes brings efficiency and data-driven results, but also risks.  Machine learning creates models which make predictions based upon patterns learned from past data. The reasoning behind these decisions is not available to the users of the models, or recipients dealing with the consequences of the decisions. E.g., a sales person doesn't know why a business is a good lead, and credit doesn't know why credit is denied.  This is a case study on adding explanations to machine learning algorithms, so that users will have greater confidence and insight into machine-driven decisions.

Session description
Speaker
Rob RollestonPaychex
Manager, Data Science
Paychex
PAW Healthcare Case Study: Healthcare Applications of Natural Language Processing

Dr. Talby will review case studies from real-world projects that built AI systems that use Natural Language Processing (NLP) in healthcare. These case studies cover projects that deployed automated patient risk prediction, automated diagnosis, clinical guidelines, and revenue cycle optimization. He will also cover why and how NLP was used, what deep learning models and libraries were used, and what was achieved. Key takeaways for attendees will include important considerations for NLP projects including how to build domain-specific healthcare models and using NLP as part of larger machine learning and deep learning pipelines.

Session description
Speaker
David Talby Ph.D
Chief Technology Officer
Pacific AI
PAW Industry 4.0
Case study: Uber

In today’s digital age, users expect a fast, reliable mobile experience. Degradations (also referred to as regressions) in mobile app performance affect not only user experience, but even hurt business metrics. However, existing mobile app release pipelines lack the necessary infrastructure to detect regressions in a mobile app's performance before it is rolled out to the world. At Uber, we are building a state-of-the-art mobile regression detection pipeline, with the goal to detect regressions as small as 1%. Our approach includes both technological innovation as well as employing machine learning along with statistical testing techniques to improve the sensitivity of the regression experiments.

Session description
Speakers
Ritesh AgrawalUber
Senior Data Scientist
Uber
Anando SenUber
Product Manager
Uber
Deep Learning World

On the forefront of deep learning research is a technique called reinforcement learning, which bridges the gap between academic deep learning problems and ways in which learning occurs in nature in weakly supervised environments. This technique is heavily used when researching areas like learning how to walk, chase prey, navigate complex environments, and even play Go. In this session, Martin Görner will detail how a neural network can be taught to play the video game Pong from just the pixels on the screen. No rules, no strategy coaching, and no PhD required. Martin will build on this application to show how the approach can be generalized to other problems involving a non-differentiable steps that cannot be trained using traditional supervised learning techniques.

This is a prelude to Martin’s full day workshop on Thursday, June 20th: Hands-On Deep Learning in the Cloud: Fast and Lean Data Science with Tensorflow, Keras, and TPUs.

Session description
Speaker
Martin GornerGoogle
Developer Relations
Google
10:50 am
Exhibits & Morning Coffee Break
Exhibits & Morning Coffee Break
Exhibits & Morning Coffee Break
Exhibits & Morning Coffee Break
Exhibits & Morning Coffee Break
11:20 am
PAW Business Track 1: BUSINESS - Analytics operationalization & management
Professional services; project management
11:20 am - 11:40 am
Case Study: Wells Fargo

The age of extending consulting services to help firms find analytical insights in their data is coming to a close. As businesses and institutions become more savvy in mining their own data, the traditional insights generated from consulting services (both internal and external) is moving to a new paradigm - data science products. This talk explores this shift in the industry and what it means for analytics and data science professionals, including how rapid advances in machine learning and artificial intelligence technologies are necessitating changes in how we think about project management, professional services, and analytics delivery models.

Session description
Speaker
Nathan SusanjWells Fargo
Vice President, Data Science Manager
Wells Fargo
Predictive Analytics World for Business
Track 1: BUSINESS - Analytics operationalization & management
Predictive analytics management
11:45 am - 12:05 pm

Innovative and impactful data science work happens when there is a mix of talented data science professionals, challenging business problems and (most importantly) data. In order to build data science solutions at scale however, the data fueling the analytical work must be clean and easily accessible to the advanced algorithms that will be leveraging it. This presentation will cover how the critical tasks of data acquisition, cleaning, storage and pipeline development must be considered when designing and operationalizing large scale data science solutions.

Session description
Speaker
Bob BressComcast Cable
Vice President of Analytics & Business Intelligence
Freewheel, A Comcast Company
PAW Business Track 2: DATA - Data Strategies & data prep
Data preparation
Case Study: Western Seminary (student retention)

One of the biggest challenges in corporations is the training of new data scientists to build the most predictive models possible with a given data set and modeling algorithm. Following the approach he's developed teaching this critical topic area after more than 20 years of industry practice, Bob Nisbet will demonstrate  the effectiveness in preliminary models of using a progressive series of common data preparation steps -- each on the same data set (KDD-Cup 1998 data set) -- including:

  • Filling of missing values
  • Derivation of "dummy variables"
  • Feature selection
  • Deriving custom variables, based on business insights, which become powerful predictors
  • Showing how to incorporate time-series data as predictors of system response with a given prediction horizon
  • Showing how different data conditioning operations (e.g. balancing and standardization) can generate very different predictive outcomes
Session description
Speaker
Bob NisbetUniversity of California, Irvine Extension
Instructor
University of California, Irvine
PAW Business Track 3: CASE STUDIES - Cross-industry business applications of machine learning
Restaurant analytics; cross-organizational data sharing
11:20 am - 11:40 am
Case Study: Stacked Restaurants

The restaurant industry in America is closing on $800b in annual revenue.  We have more than a million locations, and we employ more than 14.7m employees*. But for all of that, 9 out of 10 of our managers started at the entry level. Ex-dishwashers, busboys, and hosts, now helping us to run an $800b a year business.  

Session description
Speaker
Brian PearsonStacked Restaurants
CIO
Stacked Restaurants
Predictive Analytics World for Business
Track 3: CASE STUDIES - Cross-industry business applications of machine learning
Legal applications
11:45 am - 12:05 pm

Every year, corporations spend more than $250B on litigation in the US. The critical decisions - whether to litigate or settle or where to file suit – are often made the same way they were 100 years ago. To gain insight that companies could use to make informed decisions on legal proceedings, we built a predictive analytics engine. The approach, combining minimal viable prediction with data from thousands of patent appeal cases over 10 years, was developed to predict outcomes in future patent appeal cases. We think of it like Moneyball, but for a market 20x the size of Majors.

Session description
Speaker
Jonathan KleinEx Parte
Founder & CEO
Ex Parte
PAW Financial
Model interpretability
Case Study: Experian

Predictive models are increasingly used for important decisions such as which customers may open a financial account. These decisions affect the opportunities available to customers and drive business results. To maintain the customers' trust, it is important to be able to explain individual predictions. Likewise, it is important to explain the model logic for business managers, compliance professionals and regulators who expect fair decisions.  This is not easy when using advanced techniques such as ensemble models.  Mr. Duke will share Experian's recent advances in explainable AI technologies, with results in credit risk modeling, synthetic identity detection and fraud prevention.

Session description
Speaker
Brian DukeExperian
Data Modeling Director
Experian
PAW Healthcare

Healthcare has always used statistical analysis and analytic capabilities for accounting, reimbursement, actuarial and fiscal projection purposes. New developments in advanced statistical and predictive analytics techniques promise to revolutionize health and medical outcomes, and care delivery. These new techniques utilize modern machine learning and Artificial Intelligence methods to predict and prescribe at the individual level, instead of using traditional statistics. Learn how new machine learning techniques are being used for value-based purchasing, population health, healthcare consumerism and precision medicine. Peer into the future of Healthcare Data Science with predictions from industry leaders.

Session description
Speaker
Ken Yale, JD, DDSUCI, Irvine
Instructor
University of California - Irvine
PAW Industry 4.0 KEYNOTE
Case study: GM

At GM we are committed to have a world with zero crashes, zero emissions and zero congestion. The causes of crashes are directly correlated to human intelligence, that is, learning from experience, adapting to new situations, and using knowledge to prevent accidents.

Accident Detection and Avoidance Systems (ADAS) are a big step taken by car manufacturers to prevent the accidents. The effects of ADAS on the human attention and perception differ depending on the road conditions as well as areas (i.e., rural vs urban roads).

In a pair of case studies, we examine ADAS systems and insurance underwriting risk.

Session description
Speakers
Meltem Ballan Ph.DGeneral Motors - GM
Chief Data & Analytics Office
General Motors
Robert WelbornGeneral Motors - GM
Head of Data Science, Connected Car
General Motors
Deep Learning World

In 2015 for the first time, we demonstrated the application of the deep neural networks to the prediction of the chronological age of the patient using the basic anonymized clinical test data and launched the aging.ai system for public testing. Since then we demonstrated the integration of multi-modal data for aging research by launching the intelligently-formulated nutraceuticals and establishing a real-world data collection effort with the launch of the young.ai system. The talk will focus on the predictors of chronological and biological age using the deep neural networks trained on the blood biochemistry, transcriptomic and imaging data.

Session description
Speakers
Polina Mamoshina
Senior Research Scientist
Insilico Medicine
Alex Zhavoronkov
CEO
Insilico Medicine
12:05 pm
Lunch
Lunch
Lunch
Lunch
Lunch
1:10 pm
PAW Business Special plenary session:

The core Bayesian idea, when learning from data, is to inject information — however slight — from outside the data.  In real-world applications, meta-information is clearly needed — such as domain knowledge about the problem being addressed, what to optimize, what variables mean, their valid ranges, etc.  But even when estimating basic features (such as rates of rare events), even vague prior information can be very valuable. This key idea has been re-discovered in many fields, from the James-Stein estimator in mathematics and Ridge or Lasso Regression in machine learning, to Shrinkage in bio-statistics and “Optimal Brain Surgery” in neural networks.  It’s so effective — as I’ll illustrate for a simple technique useful for wide data, such as in text mining — that the Bayesian tribe has grown from being the oppressed minority to where we just may all be Bayesians now.

Session description
Speaker
John Elder Ph.D.Elder Research
Founder & Chair
Elder Research
PAW Financial Expert Panel:

In the insurance and banking industries, the track record of contributions made by women continues to grow. This is helping pave the way for future female scientists and analytics leaders. Predictive analytics and machine learning are no exception. At this panel session, learn from women in these fields what they've learned along the way, their wins and losses, and how they are helping others do the same. Our expert panelists will address questions such as:

  • How can you best fit in and stand up as a woman in predictive analytics and machine learning?
  • What are the key elements of being successful women scientists in these fields?
  • What are the key elements of being successful women analytics leaders?
  • How can you best build and manage your analytics team as a female analytics leader?
  • How can you increase the count of women in your analytics team, especially in leadership roles?
  • What are the differences from other science and engineering fields in terms of male domination?
  • How do you suggest balancing work and personal life?
Session description
Moderator
Mei Najim
CSPA, Founder and Lead Data Scientist
Advanced Analytics Consulting Services, LLC
Panelists
Shingai ManjengwaFireside Analytics
Chief Executive Officer
Fireside Analytics Inc.
Zeydy Ortiz
CEO
DataCrunch Lab
Jennifer Lewis PriestleyKennesaw State University
Professor of Applied Statistics and Data Science
Kennesaw State University
Anne G. RobinsonKinaxis
Chief Strategy Officer
Kinaxis
PAW Healthcare Expert Panel
The Session Description will be available shortly.
Session description
PAW Industry 4.0 Special plenary session

The core Bayesian idea, when learning from data, is to inject information — however slight — from outside the data.  In real-world applications, meta-information is clearly needed — such as domain knowledge about the problem being addressed, what to optimize, what variables mean, their valid ranges, etc.  But even when estimating basic features (such as rates of rare events), even vague prior information can be very valuable. This key idea has been re-discovered in many fields, from the James-Stein estimator in mathematics and Ridge or Lasso Regression in machine learning, to Shrinkage in bio-statistics and “Optimal Brain Surgery” in neural networks.  It’s so effective — as I’ll illustrate for a simple technique useful for wide data, such as in text mining — that the Bayesian tribe has grown from being the oppressed minority to where we just may all be Bayesians now.

Session description
Panelist
John Elder Ph.D.Elder Research
Founder & Chair
Elder Research
Deep Learning World
Case Study: Uber AI

In this talk, Chandra Kahtri, Senior AI Scientist at Uber AI, formerly at Alexa AI, will detail various problems associated with Conversational AI such as speech recognition, language understanding, dialog management, language generation, sensitive content detection and evaluation and the advancements brought by deep learning in addressing each one of these problems. He will also present on the applied research work he has done at Alexa and Uber for the problems mentioned above.

Session description
Speaker
Chandra KhatriUber
Senior AI Scientist
Uber AI
2:00 pm
PAW Business Sponsored Session
The Session Description will be available shortly.
Session description
Sponsored by
Cicero Group US
PAW Financial Sponsored Session
The Session Description will be available shortly.
Session description
PAW Healthcare Sponsored Session
The Session Description will be available shortly.
Session description
PAW Industry 4.0 Sponsored Session
The Session Description will be available shortly.
Session description
Deep Learning World Sponsored Session
The Session Description will be available shortly.
Session description
2:10 pm
5-minute transition between sessions
5-minute transition between sessions
5-minute transition between sessions
5-minute transition between sessions
5-minute transition between sessions
2:15 pm
PAW Business

The companies getting the most value from advanced analytics spend much more of their time and money embedding analytics into their core workflows than others. The most successful, in fact, spend more than half their analytics budget not to build analytics, but to deploy and operationalize it. Companies that don’t complete this last mile, those that stop once they have completed the core analytics, see their analytic investments go to waste. Join this expert panel to hear what you can do to make sure you can embed analytics in your front line and maximize the return on your analytics investment.

Session description
Moderator
James TaylorDecision Management Solutions
CEO
Decision Management Solutions
Panelists
Ryohei Fujimaki
CEO
dotData
Krishna Kallakuridiwo
President, diwo – Cognitive Decision Making Platform
diwo
PAW Financial
Model integration and deployment
2:15 pm - 2:35 pm
Case Study: Gallagher Bassett

"Build a better mousetrap and the world will beat a path to your door." Build a dozen mousetraps, with different triggers, bait, and alarms all meant for different species of mice, and the world will beat your door down. This presentation addresses the challenges of alert fatigue generated by successful predictive models. When your target audience is presented with multiple models recommending different and occasionally overlapping but important actions, there needs to be harmony in the message sent. This presentation examines the interactions of prioritization, frequency, severity development, consistency, messaging, and user socialization across multiple models for effective action.

Session description
Speakers
Gary AnderbergGallagher Bassett
SVP of Claim Analytics Product Management
Gallagher Bassett
Theodore ZitonGallagher Bassett
Vice President - Risk Analytics
Gallagher Bassett
Predictive Analytics World for Financial
Analytics start-ups; insurance applications
2:40 pm - 3:00 pm
Case Study: MotionAuto

In launching the insure-tech start-up MotionAuto, we destroyed industrial norms in the time and cost taken for reporting, analytics and decision management. In this session, we'll cover our process for:

  • Bringing data, analytics and their insights closer to the business leadership
  • Accelerating and enhancing decision making through real time access, self service and "Ask an expert" capabilities
  • Applying new knowledge insights and enhancements through a CI/CD framework.
Session description
Speaker
Robert LakeMotionAuto
Senior Manager, Data Science
MotionAuto
PAW Healthcare
Case Study: Answering One of the Big Questions

How much data is enough to build an accurate model?  This is often one of the first and most difficult questions to answer early in any machine learning project.  However, the quality and applicability of your data are more important considerations than quantity alone.  This talk presents some insights and lessons learned for gauging the suitability of electronic health record (EHR) training data for a desired project.  You will see how to determine if more data might increase accuracy and how to identify any weaknesses a model might have as a result of your current training data.

Session description
Speaker
Jeff HeatonRGA
VP, Data Scientist
Reinsurance Group of America
PAW Industry 4.0

The companies getting the most value from advanced analytics spend much more of their time and money embedding analytics into their core workflows than others. The most successful, in fact, spend more than half their analytics budget not to build analytics, but to deploy and operationalize it. Companies that don’t complete this last mile, those that stop once they have completed the core analytics, see their analytic investments go to waste. Join this expert panel to hear what you can do to make sure you can embed analytics in your front line and maximize the return on your analytics investment.

Session description
Moderator
James TaylorDecision Management Solutions
CEO
Decision Management Solutions
Panelists
Ryohei Fujimaki
CEO
dotData
Krishna Kallakuridiwo
President, diwo – Cognitive Decision Making Platform
diwo
Deep Learning World
Case Study: Microsoft

In applications like fraud and abuse protection, it is imperative to use progressive learning and fast retraining to combat emerging fraud vectors. However, somewhat unfortunately, these scenarios also suffer from the problem of late-coming supervision (such as late chargebacks), which makes the problem even more challenging! If we use a direct supervised approach, a lot of the valuable sparse supervision signal gets wasted on figuring out the manifold structure of data before the model actually starts discriminating newly emerging fraud. At Microsoft we are investigating unsupervised learning, especially auto encoding with deep networks, as a preprocessor that can help tackle this problem. An auto-encoding network, which is trained to reconstruct (in some sense) the input features through a constriction, learns to encode the manifold structure of the data into a small set of latent variables, similar to how PCA encodes the dominant linear eigen spaces. The key point is that the training of this auto-encoder happens with the abundant unlabeled data – it does not need any supervision. Once trained, we then use the auto-encoder as a featurizer that feeds into the supervised model proper. Because the manifold structure is already encoded in the auto-encoded bits, the supervised model can immediately start learning to discriminate between good and bad manifolds using the precious training signal that flows in about newly emerging fraud patterns. This effectively improves the temporal tracking capability of the fraud protection system and significantly reduces fraud losses. We will share some promising early results we have achieved by using this approach.

Session description
Speaker
Anand OkaMicrosoft
Principal Program Manager Lead
Microsoft
3:00 pm
Exhibits & Afternoon Break
Exhibits & Afternoon Break
Exhibits & Afternoon Break
Exhibits & Afternoon Break
Exhibits & Afternoon Break
3:30 pm
PAW Business Track 1: BUSINESS - Analytics operationalization & management
Training your analytics team
3:30 pm - 3:50 pm
Case Study: LinkedIn

Talent is hard to come by. The state of the art in data analytics is moving too quickly for hiring only from a pool of 4-year degree grads. At the same time, bridging the skills gap is crucial for the success of your data team and everyone who works with that team. As well, what you need your analytics team to know is growing exponentially: To truly contribute, they need soft skills (e.g. communication, conceptual understanding of vertical tech topics, the “What") as well as hard skills (the How and Why of vertical tech topics).

LinkedIn--a company with data science at its core—is working to make all stakeholders fluent in analytics. We’ve begun deploying in-house data science training, and are working to adapt these courses for more generalized public usage and benefit, by releasing them on the LinkedIn Learning platform.

In today’s session, we’ll take a look at the rationale, strategic problem solving, and pedagogical issues behind the development of the LinkedIn AI Academy: lessons learned and practical takeaways for the PAW audience. 

Session description
Speaker
Steve WeissLinkedIn
Content Manager, Data Science and Business Analytics
LinkedIn
Predictive Analytics World for Business
Track 1: BUSINESS - Analytics operationalization & management
Analytics team building
3:55 pm - 4:15 pm
Case Study: Facebook

Analytics has become extremely valuable as it enables businesses to analyze their data and drive data driven decisions by uncovering insights and predicting outcomes. In this talk, I will share my personal story on how to hire, build and maintain world-class analytics teams. 

Session description
Speaker
Nav KesherFacebook
Head of Data Sciences, Facebook Marketplace Experience
Facebook
PAW Business Track 2: DEPLOYMENT - Predictive model deployment & integration
Deployment and integration
3:30 pm - 3:50 pm
Case Study: BB&T

It's been said that Data Science is 80% data preparation and 20% modeling - but what happens when the model is done?  Where do all the models go when they're approved?  And how do we know they're still working as expected - 1 month, 6 months or a year later? 

In this session, we'll explore common challenges in model deployment, version control, and ongoing performance monitoring. Examples will cover best practices in applying DevOps methods to model deployment and automating your Data Science pipeline, including the use of GitLab.

Session description
Speaker
Alexander MoyBB&T
AVP, Enterprise Analytics & Data Science
BB&T
Predictive Analytics World for Business
Track 2: DEPLOYMENT - Predictive model deployment & integration
Deployment and model management
3:55 pm - 4:15 pm
Case Study: Canada Post

Today's organizations have billions of dollars riding on the accuracy and performance integrity of analytical models. With model performance becoming a strategic enabler and a potential source of liability, organizations need to manage the risks associated with analytics. 

To manage these risks effectively and move beyond simple financial model or spreadsheet auditing, organizations need a system of controls around analytic model development. These analytics controls provide checks and balances around model selection, validation, implementation, and maintenance.

Session description
Speaker
Allan SammyCanada Post
Director, Data Science and Audit Analytics
Canada Post
PAW Business Track 3: CASE STUDIES - Cross-industry business applications of machine learning
Entertainment industry applications
Case Study: Sony Pictures

British Prime Minister Benjamin Disraeli once said, "There are three kinds of lies: lies, damned lies, and statistics." Hollywood is gradually coming around to data-driven decision-making, but some skepticism towards quantitative analysis still lingers on. This presentation will provide an overview on movie-related metadata and how data silos are starting to break down at studios. Additionally, AI/machine learning entertainment industry examples will be shared to show how an improving analytics culture is providing actionable insights to: (1) mitigate risk when green-lighting movies, (2) improve box office predictions by building better statistical models, and (3) drive profits with targeted marketing campaigns.

Session description
Speaker
Timothy ParkSony Pictures Entertainment
Director Data Management
Sony Pictures Entertainment
PAW Financial
Fraud detection
3:30 pm - 3:50 pm
Case Study: Enova

While many businesses understand the value of advanced analytics in decision making, operationalizing data science can be challenging. This session reveals how Enova International has been successful at integrating traditional operations with advanced analytics to turn fraud defense into a collaborative analytics function. Over time, through combining the latest technologies in data, machine learning and decision automation with manual investigations, Enova International has been able to attract and retain top analytics talent, mitigate fraud risk, improve profitability and deliver a better customer experience.

Session description
Speaker
Joe DeCosmoEnova International
Chief Analytics Officer
Enova International
Predictive Analytics World for Financial
Lifetime value modeling
3:55 pm - 4:15 pm
Case Study: Institute of Consumer Money Management

Many businesses determine customer lifetime value (CLTV) in order to plan how to attract and retain customers.  Traditionally, they use descriptive analytics to determine the average CLTV.  However, with the expectation of receiving personalized services, these methods are inadequate.  Predicting how long a new customer is expected to stay as a customer, and consequently their expected CLTV, companies can make decisions on the best way to serve them.  In this talk, we will discuss practical tips and lessons learned in building machine learning models for determining CLTV including pitfalls to avoid and how deployment affects the model selection.

Session description
Speaker
Zeydy Ortiz
CEO
DataCrunch Lab
PAW Healthcare
3:30 pm - 3:50 pm

With the advent of big data and machine learning, there is an opportunity to combat rising healthcare costs by leveraging data in an ethical and privacy compliant way to establish more consistency and implantation of preventative care. We need to ensure there is a fundamental set of rules and responsibilities in place among healthcare organizations to protect their patient's privacy. In this presentation we will address this challenge and speak to the importance of creating an ethical and privacy compliant approach to aggregating multiple data sources which then can be used to improve patient outcomes.

Session description
Speakers
Casey GentryAcxiom
Senior Manager Engineering/Analytics
Acxiom
Jamie NethertonAcxiom
Senior Analyst
Acxiom
Predictive Analytics World for Healthcare
3:55 pm - 4:15 pm
Case Study: Significant Event Reduction

Failure to recognize patient deterioration is a significant issue for acute care hospitals. Code blues occur when a patient enters cardiac arrest, with only a 17-25% of leaving a hospital alive.

Hamilton Health Science (HHS) leveraged ThoughtWire’s EarlyWarning application solution to speed up response times to code blues to be 8.8 times faster than previously possible. They also reduced instances of code blues by 61%.

This session will aim to educate attendees on how Operations Performance Management (OPM) can generate real results for hospitals. Attendees will also learn about key considerations when looking to orchestrate IoT systems and devices in the healthcare setting.

Session description
Speaker
Michael MonteithThoughtWire
Chief Executive Officer & Co-Founder
ThoughtWire
PAW Industry 4.0
Case study: Vistra Energy

Vistra Energy is one of the largest energy companies in the United States, owning both power generation and retail operations in extremely competitive markets throughout the country. We combine the big data opportunities we have available as a utility with advanced analytics to provide premium customer services that differentiate us from other utility providers. We will present three case studies demonstrating how we are able to leverage our firehose of 15-minute interval IOT device energy usage data from over 1.5 million customers with advanced modeling techniques to provide added value to our customers and increase brand loyalty.

Session description
Speaker
Greg StevensVistra Energy
Manager, Advanced Analytics
Vistra Energy
Deep Learning World 3:30 pm - 3:50 pm
Case Study: Collective Sense

Logs are a valuable source of data, but extracting knowledge is not easy. To get actionable information, it frequently requires creating dedicated parsing rules, which leaves the long-tail of less popular formats. Widely applying real-time pattern discovery establishes each log as its own event of a given type (pattern) with specific properties (parameters). This application makes it a tremendous input source for Deep Learning algorithms that filter out noise and present what’s most interesting. This talk reviews real-life cases where these techniques allowed to pinpoint important issues, and highlights insights on how best to elevate DL in the development lifecycle.

Session description
Speaker
Przemek Maciołek, PhD
VP of Research & Development
Collective Sense
Deep Learning World
3:30 pm - 3:50 pm
3:55 pm - 4:15pm
Case Study: BMO Financial Group

Automated modeling is already in focus by practitioners. However, applications for marketing campaigns require significant effort in data preparation. To address this bottleneck, the robotic modeler integrates a front layer, which automatically scrolls executed campaigns and prepares data for modeling, with a machine learning engine. It enables for automated campaign backend modeling, generates scoring codes, and produces supportive documentations. The robotic modeler supports generalized deep learning assembling business targets and features. Systematically running the robotic modeler provides additional benefits including perceiving input feature importance from various campaigns or estimating cross-campaign effects. It empowers “hyper-learning” derived from campaign modeling.

Session description
Speaker
Alex GlushkovskyBMO Financial Group
Principal Data Scientist
BMO Financial Group
4:15 pm
5-minute transition between sessions
5-minute transition between sessions
5-minute transition between sessions
5-minute transition between sessions
5-minute transition between sessions
4:20 pm
PAW Business Track 1: BUSINESS - Analytics operationalization & management
Analytics team building
4:20 pm - 4:40 pm

In the future, there will not be a shortage of doctors, lawyers, teachers, and accountants, there will be a shortage of people in those fields that can speak to technology. From cloud computing to mobile and social media, there is an explosion of data from technology and there is value trapped in siloed organizations where only a hand full of specialized people are empowered with the necessary skills to realize the full potential of data. The solution is to use customized and compelling, case studies to foster a practical understanding of data analytics. This talk will provide practical steps on how to build data science skills across different functions and disciplines in your organization.

Session description
Speaker
Shingai ManjengwaFireside Analytics
Chief Executive Officer
Fireside Analytics Inc.
Predictive Analytics World for Business
Track 1: BUSINESS - Analytics operationalization & management
Analytics management
4:45 pm - 5:05 pm
Case Study: Lyft

Success in a data-driven world means empowering teams with science to improve decision making through confident, replicable and trainable programs that can engage an entire organization. Analytics teams that use a scientific approach to answer business questions will accelerate actionable insights and improve user experiences.

Peter and Martin will discuss their experience driving value in organizations including Lyft, Citrix, Alibaba and Bell where data science methods for growth and insights are at the forefront of the business. Data science is a team sport, the people in the business closest to the data often are in a position to know it best. Fostering an analytic mindset throughout the organization and training teams in a scientific approach to attack the problems they encounter will produce a needed competitive advantage.

Gain speed and agility in modeling solutions to the questions in your organization for a deeper understanding of the business landscape.  

Session description
Speakers
Martin Frigaard
App Developer
Intricity
Peter SpanglerLyft
Data Scientist
Lyft
PAW Business Track 2: DEPLOYMENT - Predictive model deployment & integration
Model Deployment
4:20 pm - 4:40 pm

There is a lot of information and best practices available so data scientists can build analytic models, but much less about how analytic models can best be integrated into a company's products, services or operations, which we call analytic operations.  We describe three frameworks so that a company or organization can improve its analytic operations and explain the frameworks using case studies.

Session description
Speaker
Robert Grossman
Managing Partner
Analytic Strategy Partners LLC
Predictive Analytics World for Business
Track 2: DEPLOYMENT - Predictive model deployment & integration
Model robustness
4:45 pm - 5:05 pm

Many organizations utilize predictive models to make decisions but what happens when those models fail to deliver, or worse, are totally off? Having had to audit numerous models across diverse industries as an advanced analytics management consultant, Stephen Chen shares personal WTF experiences and distills the perils inherent in predictive modeling which are typically glossed over in data science courses and texts. 

Using real world datasets to illustrate these issues, this session aims to help stakeholders better assess the suitability of models for decision-making, as well as helping practitioners think through their datasets and processes to build more robust models.

Session description
Speaker
Stephen ChenRoyal Bank of Canada (RBC)
Director of Strategic Analytics
Royal Bank of Canada
PAW Business Track 3: CASE STUDIES - Cross-industry business applications of machine learning
Marketing analytics
Case Study: UBM (events industry)

In the event industry, use of machine learning is not commonplace. This talk is on how UBM/Informa uses automated machine learning (AML) technology to improve their sales and marketing processes. This includes application areas such as identifying the most suitable marketing plan to maximize ROI, and forecasting the number of event pre-registrants. We employed an AML platform employed  to build and deploy accurate machine learning models quickly. Informa is a leading business intelligence, academic publishing, knowledge and events business. 

Session description
Speaker
Dave ChanInforma / UBM
Vice President, Digital Innovation & Business Intelligence
Informa / UBM
PAW Financial
Text analytics; sentiment analysis
4:20 pm - 4:40 pm
Case Study: Citigroup

Topic discovery, contextual categorization, entity linking, and sentimental analysis are some of the more highlighted examples of NLP and text mining applications in banking. Recent industry developments have  focused on predictively categorizing a potential complaint, dispute, or sales practice issue, and suggesting next best actions. Although a variety of tools and techniques are available, a success  heavily relies on customized handling through contextual understanding. As we'll cover in this session, real conversation experiences with customers in chabot and robotic process automation rely heavily on the maturity of contextual understanding.

Session description
Speaker
Yulin Ningciti
Senior Director in Global Decision Management
Citigroup
Predictive Analytics World for Financial
Insurance applications
4:45 pm - 5:05 pm

Predictive analytics has been a buzzword for a few years now. It has seen success for a wide range of applications. Within the Insurance industry, several applications have emerged, across underwriting, claims, marketing, and beyond. In this session, we will highlight examples of success factors that help sustain predictive analytics in a workers compensation insurance environment.

Session description
Speaker
Munish AroraCalifornia State Compensation Insurance Fund
Senior Director
California State Compensation Insurance Fund
PAW Healthcare

Independent Pediatricians typically maintain daily patient volumes of 20-30 patients to keep their practices viable. Pediatricians also schedule appointments up to a year in advance, leading to as many as 15% of patients not showing up for appointments each day. The financial and clinical impact of these gaps in pediatric appointment books is substantial.

PCC and Rexer Analytics analyzed pediatric no-show patterns to identify the variables that truly affect appointment truancy. These insights were translated into interventions to reduce patient truancy. We present pediatric no-show patterns, key predictors, and the results several Pediatric practices are seeing with targeted interventions.

Session description
Speakers
Chip Hart
Physician’s Computer Company
Karl RexerRexer Analytics
President
Rexer Analytics
PAW Industry 4.0 4:20 pm - 4:40 pm

Acronyms abound in the area of predictive analytics and machine learning is no exception. The discipline  of predictive analytics has been used by businesses since the end of World War II. But machine learning has been at the core of this activity  since its very early business applications. In these very early days, the "machine"  itself and not the human was identifying the predictive algorithms that would optimize a given business solution albeit in a more simplistic manner.  With the advent of Big Data, this concept of machine learning  has now expanded to more complex forms

Session description
Speaker
Richard BoireEnvironics Analytics
Senior Vice President
Environics Analytics
Predictive Analytics World for Industry 4.0
4:20 pm - 4:40 pm
4:45 pm - 5:05 pm

Due to a continuing influx of IoT devices, industrial businesses today are collecting more data than ever before. Yet only 8% of businesses are using more than 25% of the data they collect. In this session, IoT expert Dave McCarthy will describe how data science services can help organizations optimize IoT data and achieve deployments that support predictive capabilities. He will walk attendees through how they can leverage data science services to verify their optimal IoT use case and operationalize their data, enabling the speed, connectivity, and accuracy they need for condition-based maintenance, data-driven diagnostics, predictive failure and more.

Session description
Speaker
Dave McCarthyBSquare
Vice President, IoT Solutions
BSquare
Deep Learning World 4:20 pm - 4:40 pm

Deep learning models have shown great success in commercial applications such as self driving cars, facial recognition and speech understanding. However, typically these models require a large amount of labeled data, presenting significant hurdles for AI startups faced with a lack of data, funding and resources. In this session, I will discuss how to overcome the cold-start problem of deep learning by using transfer learning, synthetic data generation, data augmentation and active learning. This talk will go through a real use case of invoice processing and information extraction, which is a critical step in the Account Payable process.

Session description
Speaker
Kunling GengDecision Engines Inc.
Lead Data Scientist & AI Architect
Decision Engines Inc.
Deep Learning World
4:20 pm - 4:40 pm
4:45 pm - 5:05 pm

In the talk, I will give a detailed example how a seamlessly integrated, distributed Spark + Deep Learning system can reduce training cost by 90% and increase prediction throughput by 10X. With such a powerful tool in hand, a data scientist can process more data and get more data insight than a team of 20 data scientists with traditional tools.

Session description
Speaker
Luming Wang
Head of Data
Millennium Management
5:05 pm
End of second Conference Day
End of second Conference Day
End of second Conference Day
End of second Conference Day
End of second Conference Day
CloseSelected Tags:
Share This

Get Predictive Analytics World news and event information delivered straight to your inbox.