Data Science in Action - Building a Predictive Churn Model (openSAP)

The recruitment company Glassdoor recently revealed the 50 best jobs in America. The score is determined by three key factors: the number of job openings, the job satisfaction rating, and the median annual base salary. Unsurprisingly, the data science jobs claimed the top spot for the second year in a row. In fact, the data science job score was 4.8 out of 5, with a job satisfaction score of 4.4 out of 5, and a median base salary of $110,000. One of the main reasons for these extraordinary scores is the huge shortage in talented individuals with the right skill sets.

Following the success of the openSAP course Getting Started with Data Science, this follow-on course will provide students with the knowledge and hands-on skills to build, test, and deploy predictive models in a business-like environment.

In the first week, you’ll be introduced to the business case study where you are asked to investigate customer churn for a telecommunications organization. In the second week, you’ll prepare the data and create an analytical data set, conduct an initial data analysis, and learn how to encode the data. In the third week, you’ll develop a churn model, evaluate its performance, and learn how to productionize it. Then, in the fourth week, you’ll learn how to monitor and maintain a model, build a segmentation, and improve model performance. Finally, in week five, you’ll take a short exam.

Data science is a complex subject, but in this course, you’ll revisit the fundamental principles and learn how to do if for real. Remember, you don’t have to be a rocket scientist to be a data scientist.

Course Content

Week 1: Case Study Introduction

Week 2: Prepare and Encode Data

Week 3: Develop, Evaluate, and Deploy Models

Week 4: Monitor Models and Improve Performance

Week 5: Final Exam

Prerequisites:

The following openSAP course is strongly recommended: Getting Started with Data Science