Agenda
Predictive Analytics World for Business Las Vegas 2018
June 3-7, 2018 – Caesars Palace, Las Vegas
This page shows the agenda for PAW Business. Click here to view the full 7-track agenda for the five co-located conferences at Mega-PAW (PAW Business, PAW Financial, PAW Healthcare, PAW Manufacturing, and Deep Learning World).
TOPICS – The sessions across this two-day, three-track conference are grouped into the following five topics:
Analytics operationalization & management
Track 1
Retaining & optimizing HR with analytics
Track 2, Day 1
Predictive modeling & machine learning methods
Track 2, Day 2
Manifold business applications of machine learning
Track 3, Day 1
Analytics for customer acquisition & retention
Track 3, Day 2
TOPICS – The sessions across this two-day, three-track conference are grouped into the following five topics:
Analytics operationalization & management
Track 1
Retaining & optimizing HR with analytics
Track 2, Day 1
Predictive modeling & machine learning methods
Track 2, Day 2
Manifold business applications of machine learning
Track 3, Day 1
Analytics for customer acquisition & retention
Track 3, Day 2
Session Levels:
Blue circle sessions are for All Levels
Red triangle sessions are Expert/Practitioner Level
Workshops - Sunday, June 3rd, 2018
Full-day: 8:30am – 4:30pm
This one day workshop reviews major big data success stories that have transformed businesses and created new markets. Click workshop title above for the fully detailed description.
Two and a half hour evening workshop:
This 2.5 hour workshop launches your tenure as a user of R, the well-known open-source platform for data analysis. Click workshop title above for the fully detailed description.
Workshops - Monday, June 4th, 2018
Full-day: 8:30am – 4:30pm
This one-day session surveys standard and advanced methods for predictive modeling (aka machine learning). Click workshop title above for the fully detailed description.
Full-day: 8:30am – 4:30pm
Gain experience driving R for predictive modeling across real examples and data sets. Survey the pertinent modeling packages. Click workshop title above for the fully detailed description.
Full-day: 8:30am – 4:30pm
Dive in hands-on with crucial data prep steps, including cleaning, missing value imputation, feature creation/selection, and sampling. Click workshop title above for the fully detailed description.
Full-day: 8:30am – 4:30pm
This one-day introductory workshop dives deep. You will explore deep neural classification, LSTM time series analysis, convolutional image classification, advanced data clustering, bandit algorithms, and reinforcement learning. Click workshop title above for the fully detailed description.
Predictive Analytics World for Business - Las Vegas - Day 1 - Tuesday, June 5th, 2018
Applied deep learning has fast become a standard tool for many industry machine learning applications. New advances in neural network techniques have opened the doors to solving problems at scale that were out of reach until recently. Because of these advances, applications such as in image recognition for self-driving cars, medical image classification, text translation, and fake news detection are both tractable and often the industry standard. In this keynote, Mike Tamir, who heads the data science teams at Uber ATG—the self-driving cars division—reveals how two key application areas of deep learning signal the broad importance of this emerging technology.
Data science, if judged as a separate science, exceeds its sisters in truth, breadth, and utility. DS finds truth better than any other science; the crisis in replicability of results in the sciences today is largely due to bad data analysis, performed by amateurs. As for breadth, a data scientist can contribute mightily to a new field with only minor cooperation from a domain expert, whereas the reverse is not so easy. And for utility, data science can fit empirical behavior to provide a useful model where good theory doesn’t yet exist. That is, it can predict “what” is likely even when “why” is beyond reach.
But only if we do it right! The most vital data scientist skill is recognizing analytic hazards. With that, we become indispensable.
DataRobot will demonstrate how innovations in Machine Learning & Artificial Intelligence can easily be leveraged to address a variety of opportunities and challenges across industries. Today, companies now have the ability to use machine learning & AI as a competitive advantage in their market, reduce operational costs, create new revenue
streams, and drive higher customer satisfaction and loyalty. DataRobot's presentation will outline how scaling AI across your enterprise will deliver top & bottom line-benefits...create truly an "AI-Driven Enterprise".
This PAW track – "BUSINESS - Analytics operationalization & management" – is kicked off by James Taylor of Decision Management Solutions, an industry leader in the operationalization of predictive models.
Many organizations still struggle to get a business return on their investment on advanced analytics. The biggest barrier? An inability to integrate analytics, especially predictive analytics, into frontline systems and business processes. Work with a number of global companies has revealed three critical success factors. By adopting a more decision-centric approach to analytics, changing the way requirements and business understanding are defined, and considering advanced analytics as one of a set of decision-making technologies, organizations can tie their analytics investments to business results and deliver the business value they are looking for.
Drawing on a unique database consisting of workforce (HRIS) data from multiple organizations, Mercer's Workforce Analytics team has applied predictive modeling to examine to what extent systematic differences exist between what millennials actually value in employment as compared to older employees. The results challenge "popular press" stereotypes about millennials at work. By identifying the predictive antecedents of employee turnover, i.e., what employees do, as opposed to relying on survey results, i.e., what millennials say they value, this research provides insight into actual behavior in a way that has practical relevance to the design of effective employee value propositions (EVPs) and total rewards.
10:30 - 10:50 am
Charter Communications (Spectrum) is the second largest cable service provider in the nation, and every investment to expand fiber network is based on maximizing the ROI. Charter makes an extensive use of Predictive Analytics based on Logistic Regression, Survival Analysis, GIS analysis to determine the best go-to prospects and assign sales territories to maximize return.
The marketing team has been able to improve campaigns' performance by over 30%, and the network construction team has been able to identify thousands of new buildings with the best ROI (with only 5% rejection rate) which were deemed unreasonable till now.
10:55 - 11:15 am
Turner has developed a machine learning algorithm that is capable of predicting granular audience targets. This innovation, called Competitive Audience Estimation, enabled advertisers/agencies to buy and execute audience targeted marketing campaigns on linear TV. This talk will overview the methodology and attained results.
11:20 - 11:40 am
Getting models and analytic products ‘over the line’ and implemented or operationalized can be challenging due to factors often outside the direct control of a modeling or analytics team. I’ll share some of the obstacles that we’ve met in our journey maturing as an Enterprise Data and Analytics organization. Some are business and organizational challenges, others are technology related. At Northwestern Mutual, we’ve successfully navigated a number of these challenges and deployed models and other analytic products for mission critical business functions. I’ll talk through some of the challenges and share our approach to overcoming them.
11:45 am - 12:05 pm
Data Science’s massive potential can only be unleashed if it’s aligned with, and empowered by, senior leadership in an organization. Even the most sophisticated, transformative and well-executed projects can be for naught if the right people don’t trust how it works, its value or haven’t developed a culture that values Data Science. But while senior leaders speak business, Data Science teams often speak in math or code, and increasingly do so as the field evolves. How can Data Science teams bridge this gap to elevate their place and impact within an organization? In this talk I will share the story of Quicken Loans’ Data Science team and how some simple but insightful best practices can build the trust, relationships and understanding with senior leaders to transform organizations.
11:20 am - 12:05 pm
Companies worldwide make massive investments to make their employees more productive. This endeavor has become more critical than ever as access to a qualified workforce is becoming increasingly challenging. While engaging users with relevant content is certainly a desired outcome, most product implementations still focus on optimizing click-through rate or engagement rather than productivity. As a workforce and management software market leader, Atlassian holds a unique position to determine how project documentation and team communication actually impact the throughput of employees. This talk addresses the intriguing correlations between the usage patterns observed in workforce collaboration tools, and the productivity of the user.
11:20 - 11:40 am
Our overarching goal was to determine the underlying phenomena that create market demand for educational programs. The central question was: Is it the type of program, or the lead to enrollment process, or some identifiable behavior of the leads, or lastly would the demographic attributes turn out to be the chief drivers of enrollment. Rather than model the likelihood of enrollment with all the available variables, we methodically determined the product or school program variables that made them attractive; then the process of enrollment and finally to the behavioral characteristics of prospective students.
11:45 am - 12:05 pm
Predictive modeling has gained popularity in studying college enrollment due to fierce competition in higher education. To make informed decisions and allocate limited resources to improve enrollment, predictive modeling has been applied to challenge and change the traditional recruitment process. This session is intended for two learning outcomes: Participants who are not familiar with predictive modeling will learn how to lay out a plan to collect and build a comprehensive data infrastructure and conduct predictive modeling. Participants who have run predictive modeling will learn how to critically examine the quality of their predictive analyses.
As data stores grow and more emphasis is being placed on understanding the relationships between people, places, and things, interest in graph technology has exploded.
Unlike relational databases, which rely on costly joins to reconstitute connections, graphs emphasize the connections themselves, putting far more value on the data between the data. Graphs link entities, along with their associated properties, directionally with edges that describe the relationships between them. And because it is possible to use multiple links, understanding temporal and spatial elements becomes far easier.
In this session, Brandy Freitas from Pitney Bowes, in conjunction with their partner ICC, will discuss graph databases and how they are charging analytics in a variety of industries. They will walk through the power of the graph, practical use cases, and take a dive into query building, machine learning, and native graph algorithms. With real life examples, including topics like sales prospecting and fraud detection, this session aims to get participants up to speed on the use of graph databases and showcase their importance in modern business decisions.
Looking for tips and pointers on how and where to use machine
learning? Look no further. In this session, Theresa Kushner - SVP of Dell EMC's
Performance Analytics Group - will cover best practices as they apply for
medium to large enterprises, including:
- Tips for multinationals to effectively apply predictive
analytics in their business – where, how and why?
- Hottest techniques in predictive analytics for enterprises
- Role of start-ups in boosting predictive analytics techniques;
applying external expertise
- How to effectively partner with your IT organization to get
the most out of machine learning
Despite enterprise investments in business intelligence and data science tools, the thirst for data and analytics is far exceeding the ability for organizations to deliver actionable insights from all their data. Most business users are stuck running static reports, while the data professionals who support them tackle the increasing backlog of ad hoc requests, struggling with operationalizing insights into the business workflow.
The new era of intelligent analytics applications will raise the adoption of analytics tools in the enterprise while closing the gap between BI and AI. Tellius will demonstrate how business users are empowered to ask questions through an easy-to-use, search-driven interface and effortlessly perform advanced analysis through automated discovery of insights powered by machine learning.
Decisions are never made in a vacuum. They are the product of how decision-makers perceive their policy, political, business, and organizational environment. Shaping their decisions means shaping their environment.
In this talk, Daniel Porter, Chief Analytics Officer of BlueLabs, will discuss his team’s approach for using data science to shape environments by influencing influencers. The approach pinpoints influencers across different industries and then dynamically scores them on their elasticity and level of influence within a given domain, so that organizations can target the most relevant influencers with ads and outreach, online and offline.
Using examples from influencer outreach for major organizations, Porter will talk about how his team's approach to identifying and scoring influencers, from Fortune 100 executives to government officials, staff, major donors, and policy elites. Through a combination of data science tools and algorithms ranging from NLP classification to uplift modeling and non-proportional sampling methods, this approach provides organizations with an unparalleled ability to achieve their brand and policy objectives.
In 2017, the HR Analytics team in Shell collaborated with Shell Global Commercial, and connected Sales and HR data to improve the way we hire, reward, and develop sales account managers. The research project, called Project Spark, developed an Individual Sales Performance Model to identify key drivers of Sales Performance. Statistical modelling techniques (e.g. general linear modelling, logistic regression analysis, multi-level modelling) were applied to understand the effects of, for example, Personality, Career Types, and Team Leadership on Sales Performance. Based on the insights, Global Commercial and HR are co-designing practical solutions to get the best out of our people.
2:40 - 3:00 pm
Comcast data scientists need to use the vast resources of our in-house data lake in order to perform ML tasks such as anomaly detection. In the schema-onread model of data lakes, a data scientist first has to find datasets of interest, infer the meaning of each attribute, and clean up the data to enable joins across datasets and to ensure equivalent data (e.g., timestamp or geo location) is represented in the same format or vocabulary. Typically 80% of her time is spent on these tasks, before she can begin identifying features of interest and actually running the ML models.
How does data governance make feature engineering easy? Imagine a data lake in which datasets are represented via peer-reviewed Apache Avro schemas, using common sub-schemas, with the semantics and format of each attribute well-documented. A metadata repository enables data discovery based on these schemas, and also stores end-to-end lineage from streaming data capture through storage in the lake. Governance also enforces schemas to be compatible when they evolve, so that data stored in different schema versions don’t need to be converted before use in feature engineering. Also, by definition, the features are already defined in the datasets in a common Avro representation. Finally, it allows the data preparation for ML to become generic and therefore re-usable, e.g. extracting from any data set a time series of a single value (i.e. a feature) according to several dimensions.
3:05 – 3:25 pm
The point of a business is to maximize profit. The best way is to always have the best price. But sometimes it makes sense to offer discounts to focus on the long term. This talk will focus on building and deploy such a tool, internationalization and dealing with a large number of Shiny users.
The opportunities and challenges created by the expanding size and definition of data has become the catalyst for many universities across the country to think differently. The evolving discipline of data science is changing the way universities approach education, research and funding. In particular, the way universities collaborate with the private sector for research and development has fundamental changed - at least when it comes to data science. And, importantly, this new private sector/university collaborative framework has, in turn, contributed to the evolution of Data Science for both parties.
In this session, Dr. Jennifer Lewis Priestley, director of one of the country's first Ph.D. program in Data Science, will talk about how the relationship between universities and private sector companies has been impacted by the evolution of data science.
3:55 - 4:15 pm
Our consulting company, Centric, was struggling with low utilization - too many employees were not on billable projects. We designed and implemented a machine learning model to help us better understand the factors driving employee utilization. Although we had only a few hundred rows of data with which to work, the insights gained were used to redesign internal business processes, resulting in a 13% gain in utilization which in turn resulted in a nearly 20% improvement in company-wide profitability. In this session, we will cover the process we took from designing the machine learning model to implementing our findings in "real life," including tips, tricks, and lessons learned.
4:20 – 4:40 pm
Engaging employees is critical for any organization. It doesn't need to be difficult, and it doesn't need to be expensive. In this case study, the Baptist Health Medical Group's Practice Optimization Team will demonstrate what it did (and what you can do) to measure employee engagement, and use statistics-based insights to take engagement to the next level.
In this session, we will share our approach for:
1) Asking the right questions,
2) Identifying engagement drivers,
3) Communicating insights to leadership, and
4) Creating team-generated action plans.
You already have the tools and budget to do what we did. We will show you how.
At Google, we developed analytical methods to evaluate the hypothesis that Universal Agents, who handle support calls across several products, would outperform Laned Agents, who are specialized to support only a single product. Here in Google's gTech group, initially operating under the assumption that Universal Agents would have higher utilization than the more prevalent Laned Agents, Google increased the number thereof, thereby creating the opportunity to collect the data needed to make this comparison. After several months, we compared the two strategies. To do so, however, required some innovation, since many factors impact agent performance. To normalize for these factors and get a clean adjusted estimate of performance, we applied a form of mixed modeling (an advanced form of ANOVA). Our findings were counterintuitive: Universal Agents under-perform Laned agents in overall cost and measures of customer satisfaction. This project corrected many misconceptions and resulted in a relative increase in Laned Agents.
Analytics is hard and getting stakeholders and executives to act and use the results from analytics may be even harder. In this presentation we argue that to turn analytics into advantage requires: (1) choosing the right analytical target to go after, (2) building credibility for the insights, (3) offering the tools that turn the insights into decision alternatives. We showcase these factors via 5 actual business case studies (each of which proved to benefit greatly from their efforts).
Predictive analytics tools hold great promise for harnessing the power of data to make empirically-based decisions with significant increases in efficiency and effectiveness. However, predictive analytics tools come under special scrutiny when used for high-stakes decision-making such as employment selection. Join us for an overview of foundational laws, theories of discrimination, and approaches for measuring adverse impact and validity in the use of predictive analytics for employment purposes.
IBM Technical Support handles over a million tickets per year and is committed to helping clients solve technical issues and creating optimal client experience. An always-on prediction system was recently developed to automatically forecast the daily client experience outcomes by ticket prior to ticket resolution. Support agents use it to prioritize their workload and take tailored actions to improve client experience of high risk tickets. In this session, I'll introduce the journey of developing the prediction system via iterations by combining classical statistical methods, machine learning algorithms, and human feedback to achieve significant improvement in model quality and user adoption.
5:30 PM to 5:50 PM
Attracting and retaining the right talent continues to be a struggle, especially in the Silicon Valley. Advances in statistical modeling have allowed us to predict and understand in depth and breadth why employees leave companies. This presentation will walk through some of the analytic techniques being used at Twitter, such as survival analysis, decision trees, etc. being used to predict employee attrition.
Predictive Analytics World for Business - Las Vegas - Day 2 - Wednesday, June 6th, 2018
What does a 90-year old company have to do with predictive analytics? Quite frankly, everything. Through predictive analytics services and solutions, Caterpillar is increasing the value of its equipment and commitment to its customers by helping them predict potential outcomes and make better business decisions. Morgan Vawter, Caterpillar’s Chief Analytics Director, will share how this legacy company is combining domain expertise with vast amounts of data and advanced mathematical techniques to help its customers build a better world through better fuel productivity, increased safety, planned downtime and other predictable efficiencies.
Rexer Analytics has been surveying analytic professionals for over a decade. In 2017, over a thousand people from around the world participated in the 8th Data Science Survey. In this PAW session, Karl Rexer will present highlights of recent survey results and discuss trends from the past decade. Highlights will include:
- Key algorithms
- Deep learning adoption and key techniques
- Challenges of self-service analytics
- Analytic software adoption
- Job satisfaction & job prospects
diwo (Data In, Wisdom Out) acts as a sixth sense, continuously working in the background to sense and quantify invisible opportunities and situations before they arise, so business users are empowered to act in time and make optimal decisions. Its uniquely developed framework harnesses cognitive computing and machine learning to proactively guide business decision making. diwo's business-first approach prioritizes business and user context, scalability, and value on day one. The new platform’s design tackles adoption issues common with other transformative initiatives. As diwo is vertical-agnostic, it is applicable to a wide range of specific business scenarios across industries. Its scalability allows for a variety of businesses to roll it out according to its own organization’s needs by integrating with existing data and analytics assets.
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.
As data science captures more attention from decision makers, translating the models from the language of the analyst into a language of the decision maker has become an important topic at conferences and in journals. It used to be that the focus on data storytelling was on visualization techniques. While this is important, as analyses become more complex, the task of interpreting the models likewise becomes more complex. Before we can decide on visualization techniques, we first need to uncover what to visualize. In this presentation, Mr. Abbott will describe ways to unravel complex descriptive and predictive models so they can be explained and visualized using machine learning models and resampling techniques.
10:05 - 10:25 am
By 2021, Digital Marketing spend will near $120B in US and will be 46% of overall marketing spend. People-based marketing is a foundation of improving targeting efficiency. Successful targeting requires a combination of sophisticated modeling, deployment and application of both think and learn and test and learn methods. In this presentation, we will go over some of the results and lessons learned by John Hancock in its Direct to Consumer (D2C) journey, that went from 0 - 60 in less than a year.
10:30 – 10:50 am
In 2017, our bank implemented a machine learning propensity model in our direct consumer campaign. As a result we saw a 40% reduction in marketing cost and a 70% increase in response rate. Building a machine learning model is hard; convincing the organization to implement it is harder. In this session, we will present a success story of how we convinced the management committee to implement a machine learning model that appears to them like a black box. As a result, we saw a culture shift; the organization is now more willing to explore data science solution in other products.
After 3 years of building a global analytics team for a 4B software company from bottoms up, we are now the go-to analytics shop for CA Technologies. We are considered a foundational piece of all transformational initiatives especially around sales, marketing and customer success. We’ve been featured in Forbes, Forrester and just received Information Magazine’s Analytics 50 Award. Hear about the key secrets and learnings from a successful analytics team leader that can help you reduce trial by error and avoid inevitable pitfalls.
A predictive model used in production has an incredible amount of manual overhead. That restricts the “real work” that a data scientist can do. If you can identify and automate that 85/90% of the work, what can be achieved? In this presentation, the elements of “model process automation” will be broken down and discussed using open source software throughout to demonstrate the logic. A real world example will be shown with Which? (the UK equivalent of Consumer Reports and Consumer Advocates) using 29,000 pricing ensemble models that create, train, build, tune and score every day totally automatically. From this example, best practices will be discussed.
11:20 - 11:40 am
With a portfolio of brands ranging from Nickelodeon to Comedy Central, Viacom distributes and markets content that appeals to all demographics across the globe. Still, finding the right digital audiences for tune-in marketing is a challenge. In this session, Viacom will share the tools and models they have built to create smarter audience targets, leveraging digital content consumption by TV viewers.
11:45 am – 12:05 pm
Caterpillar is implementing predictive models to identify customers that are most likely to purchase heavy equipment. In this session, a data scientist and an analytics manager from Caterpillar will discuss these models.
Machine learning (aka predictive analytics) only delivers value when acted upon – that is, when deployed. Only a carefully designed management process ensures that the analytics' output is pragmatically viable for operationalization, and that company operations – your internal consumers of analytics – know best how to employ the product they're consuming.
Lead by moderator James Taylor, an industry leader in the operationalization of predictive models, this expert panel explores and expands upon PAW Business' Track 1 topic, operationalization, to provide insights as to how best to execute on the functional deployment of machine learning.
An explosion in data availability from integrated data supply chains cutting across traditional boundaries, cheap cost of data storage and massive increase in computing power are expanding the universe of business problems that could be addressed by using predictive analytics in general and machine learning in particular. While the use of machine learning to understand complex patterns and identify non-linear relationships offers a great opportunity to enhance the accuracy of predictive models, risk management as domain has relatively seen lower adoption of machine learning algorithms.
In this presentation, Vivek will discuss the key drivers for using machine learning and present a case study where machine learning algorithms were successfully used to enhance the accuracy of risk models for a large global organization as it transformed its data supply chain and downstream & upstream technology infrastructure to digest large amount of data and operationalise the analytical outcomes in more agile manner.
The use of Predictive Analytics and Machine Learning is imperative across all channels and departments in Financial Services. Angoss differentiates itself by its easy-to-use visual data science platform with intuitive workflows and best-in-class Decision and Strategy Trees. We will discuss how best to utilize the Angoss Software Suite as the ultimate complete solution.
2:15 - 2:35 pm
Decision-support systems based on machine-learning algorithms can require a significant investment to build. Ensuring the data pipeline is reliable, building the right models, and hiring the right staff can be a major undertaking. This presentation will cover best practices for building a business case to deploy machine learning algorithms for large scale product development or other production-level systems.
2:40 – 3:00 pm
In this session, Robert Lanning will deliver an energetic and engaging discussion on the responsibilities of analytics professionals to ensure that the projects worked on actually accomplish meaningful change in the organization. Through the use of thought-provoking examples, the attendee will leave this session with an appreciation of the responsibility analytics professionals should bring to their projects.
Preeminent consultant, author and instructor Dean Abbott, along with Karl Rexer of Rexer Analytics, field questions from an audience of predictive analytics practitioners about their work, best practices, and other tips and pointers.
We searched for matches between 10+ million end customers and 250,000+ resellers along the B2B technology supply chain. The line of sight of our high-volume business is broad, capturing around half a billion transactions of 5+ million technology products globally. We built matching models and product recommendation engines based on statistical and machine learning approaches, including random forests and neural networks. Our deployed product offerings include lead generation, sales forecasting and cross-selling. In this session, we cover this project and show that the insights gained that pertain to feature engineering, model building and validation exercises can be scaled to industries with similar lead generation and matchmaking challenges, including Education, Healthcare and Travel.
This talk will review recent research by the International Institute For Analytics that studied the analytics maturity level of large enterprises. A range of interesting findings and trends were discovered. The talk will cover how maturity varies by industry and some of the key steps organizations can take to move up the maturity scale. The talk will also cover some exciting research that correlates analytics maturity with a wide range of corporate success metrics including financial and reputational measures. The results reinforce that organizations with more success in analytics have more success overall.
Machine learning on spatial data is hard. In this session, Peter Lenz, one of the few geographers in the machine learning world, will cover some of the unique issues when working with such data, how to treat it correctly, and why Dstillery ended up building two parallel spatial featurizers to maximize the utility of geo data for its machine learning stack.
Natural Language Processing (NLP) algorithms are a part of emerging field within Artificial Intelligence (AI) that focuses on the interactions between human language and computers. By utilizing NLP algorithms, we can organize and structure knowledge to perform tasks such as automatic summarization, translation, relationship extraction, sentiment analysis, speech recognition, and topic segmentation. We leveraged NLP algorithms to dive deeper into customer complaint narratives and to identify major complaint themes. Moreover, we quantified underlying consumer sentiment and ultimately develop actionable consumer insights. These insights were useful for developing more effective Marketing and Customer Experience strategies.
Data professionals expect a lot from their leaders and organizations, and they are in higher demand than ever. How do you build an organization that Data Scientists are proud to be a part of, one that leverages the network effects of the analytics community as a whole? How do you keep your team focused on the projects that drive organizational goals forward, while satisfying the insatiable hunger for technical knowledge and “interesting work?” And a leader’s work doesn’t stop there: How do you get the most out of the cross-functional skills that your team members possess (both technical and otherwise), and how do you ensure that the critical analytics roles in your organization are properly recognized and rewarded for their contributions? Join Cisco for an insider look at what’s working today…and what’s planned for tomorrow.
We've all heard the hype. Being Bayesian is cool, right? But does it really offer any clear benefits over the frequentist approach, or is it just a different way of thinking about the solutions we already have on hand? The Bayesian approach does enable a very powerful class of models: multi-level/hierarchical models. With a new generation of statistical modeling platforms like the free tool STAN, these models can be built with rich structures that enable more granular insights. We will learn about the power of multi-level models through three concrete examples: digital UX optimization, market response modeling, and programmatic media optimization.
Time series analysis has not benefited much from the data science halo effect compared to, say, statistical learning techniques. Yet time series model are often used to inform some of the most critical and fundamental business decisions.
Case in point: when evaluating the overall health of a business, we typically look at high-level metrics such as CAC, top-line growth and margin. And when it comes to forecasting these metrics and planning the underlying investments in marketing, inventory and promotions, you need time series analysis
In this talk we’ll take a random walk through topics related to time series analysis as it pertains to planning:
- Why having robust planning models is important, even though you’ll always be wrong
- The skill set and attitude needed to be good at building the underlying models
- Embracing uncertainty as opposed to trying to make it go away
- How to think about promo/price versus top-of-funnel spending
- When ML and time series come together
Workshops - Thursday, June 7th, 2018
Full-day: 8:30am – 4:30pm
This one-day session reveals the subtle mistakes analytics practitioners often make when facing a new challenge (the “deadly dozen”), and clearly explains the advanced methods seasoned experts use to avoid those pitfalls and build accurate and reliable models. Click workshop title above for the fully detailed description.
Full-day: 8:30am – 4:30pm
This workshop dives into the key ensemble approaches, including Bagging, Random Forests, and Stochastic Gradient Boosting. Click workshop title above for the fully detailed description.
Full-day: 8:30am – 4:30pm
This workshop demonstrates how to build uplift models (aka net lift models) that optimize the incremental impact of marketing campaigns, covering the pros and cons of various core analytical approaches. Click workshop title above for the fully detailed description.
Full-day: 8:30am – 4:30pm
Gain the power to extract signals from big data on your own, without relying on data engineers and Hadoop specialists. Click workshop title above for the fully detailed description.
