By: Daniel Egger, Executive in Residence and Director, Center for Quantitative Modeling
Last Spring, I competed for the right to develop a series of four online courses in Business Data Analytics for Coursera: what Coursera calls a “High-Demand Specialization.” We won!
The Specialization is called Excel to MySQL: Analytic Techniques for Business and it’s my primary responsibility to deliver the first two of four courses: Business Metrics for Data-Driven Companies, (which launched September 15th) and Mastering Data Analysis in Excel (launched October 19th). I’m also developing a really cool Specialization Capstone Project in collaboration with Airbnb, which will go live on January 18th, and involves student developing their own predictive models to optimize the rental value of residential properties.
Along with my collaborator, Neuroscience postdoctoral student Jana Schaich-Borg of the Information Initiative at Duke – see bigdata.duke.edu – and a team from Duke’s Center for Instructional Technology, we have been working seven days a week to deliver four Courses and Capstone on a tight schedule. Many current MEM students are contributing directly to the effort, as Teaching Assistants on one or more of the Specialization Courses, helping us catch mistakes, develop practice problems, and provide high-quality supplementary materials.
The first Course aims to be a non-technical overview of the ways data analysts, business data analysts, data scientists and other technical folk interact with data in the business world. It is organized around what business metrics are most important to track, and what kind of jobs people actually do that involve Big Data: what tools they use, and how they identify opportunities to increase revenues, maximize profits, or reduce risk. The course also explores how different types of companies, with different corporate cultures, are responding (or not responding!) to the competitive opportunity and threat of Big Data – how effectively they are embracing “Big Data Culture.”
As part of my preparation, I got in touch with a good number of MEM alumni who are pursuing careers in data-analytics fields. It was cool to hear about the exciting jobs our students are doing all over the world. Many are working with the very latest technologies and are applying them in completely new ways. It is very gratifying to see how much responsibility our students have already taken on early in their careers, how much they are obviously enjoying their work, and how well prepared they are to succeed in a Big Data world! Three of my former Data Mining students seemed so representative of the rest – and so articulate – that I decided to include interviews with them in Course 1.
Over 6,000 learners enrolled in Business Metrics for Data-Driven Companies in its first two weeks since launch. Many learners have commented in the forums on how much they like the interviews. Making the interviews was my favorite part of creating the course.
The interviewees are:
Shambhavi Vashishtha (MEM Fall 2012) who works as a Business Analyst at Opera Solutions, a leading IT-focused strategic consulting firm. Video Interview
Tiffany Ting Yu (MEM Fall 2012) at the time of the interview a Business Data Analyst at Argus Information & Advisory Services, a strategic consulting firm which has developed its own proprietary databases and specializes in helping banks market credit cards and manage their credit card risk (since September 2015 Tiffany is working in a similar role at Goldman Sachs). Video Interview
Dai Li (MEM Spring 2013) who works as a Data Scientist at If(We), a high-tech startup in the social networking arena. Video Interview
Really, these interviews speak for themselves – opportunities for MEM graduates with a strong interest in data science and a willingness to acquire new technical skills on the job are practically unlimited in today’s market.
Click Here to see the 3 full videos!
By the way, Course 2 – Business Data Analytics with Excel – launched last week! – is based closely on the Data Mining course I’ve developed over the last six years for Engineering Management students. We use a simple and accessible data-processing tool – Excel – that raises minimal technical barriers to participation – but nevertheless develop mathematically deep and generalizable (Bayesian Logical Data Analysis) methods that aim in the long run to help rationalize the field of data science.
Course 2 focuses on how quantitative measures of Information, uncertainty, and reduction in uncertainty or information gain, bring accountability to the work of data scientists. Information measures are objective and can provide a shared conceptual framework to allow all stakeholders to track the incremental value of an individual model, or an entire data-science engineering initiative, independently of the technical details of the algorithm or project.
The videos and supplemental materials we’ve created for both Course 1 and 2 will I hope also be a valuable resource for future Duke Data Science students – by covering basic principles online, we should free-up more class time for individual project work with real data sets.
Course 3 – Data Visualization & Communication with Tableau and Course 4 – Managing Big Data with MySQL – are being developed primarily by Jana. They will launch in November and December respectively. We hope you will join us for some or all of this adventure!