- Recognize some of the opportunities possible with big data and predictive analytics.
- Understand some of the problems that companies face due to the growing volume of data
- Gain experience looking at analytics from a strategic perspective
- Understand how big data and analytics is impacting IT and operations
- Learn about the new IT infrastructure designed to support big data and analytics
- Learn about some of the basic tools available to manage data and build predictive models over it.
- Learn about the structure of a typical analytic project and some of the reasons that some succeed and some fail
- Understand the role of predictive analytics in marketing
- Understand the role of predictive analytics in increasing sales
- Understand the role of predictive analytics in optimizing operations and logistics
- Understand the role of predictive analytics reducing risk and protecting revenue
About the choice of topics:
This course provides a management perspective on big data and analytics.
You can get a better feeling for the material covered by looking at the course syllabus. Although techniques to manage big data and to build statistical models over it have been around for many years, most companies do not make effective use of their data to increase their revenues, reduce their risks, and optimize their operations. The goal of the course is to introduce through a series of case studies (Capital One, Netflix, FICO, Google, Progressive, and the Stanford Hospital and Clinics) some of the successful strategies that companies and other organizations use to analyze their data, build statistical models, and deploy these statistical models in their products, services and operations. This course will not give you the technical skills to manage a Petabyte of data in Hadoop, build statistical models over it using R, and deploy these models into an enterprise architecture. On the other hand, it may provide you some insights into why some companies are better at this than others.
The course is not designed to teach you how to build statistical models. If you are interested in learning how to build statistical models you should take one of the courses that teaches statistical modeling or download the open source R system and learn to use it.
The course is also not designed to teach you how to set up a Hadoop-based system. If you are interested in learning how to set up Hadoop, you can download Hadoop and work through an introductory book on Hadoop.
The course will provide a high level introduction to some standard techniques that are used with big data, including clustering, regression and classification trees, Naïve Bayes models, and collaborative filtering. If you are already familiar with these techniques, this course is probably not a good course for you.
The course will also provide a high level introduction to Hadoop. If you are already familiar with databases and data warehouses, this course is probably not a good course for you.
A lot of what passes for new insights about big data and analytics today is not new and has been known for 40 years or longer. In this course, we try to provide a broad perspective on big data and analytics. For this reason, some of the case studies are from the 1980’s and 1990’s and earlier.