Data science for business

Summary

Students will learn how data science can support business decisions from prediction and evaluation to sequential decision-making. The course covers core machine learning tools for business analytics and introduces reinforcement learning as a framework for optimizing repeated business decisions.

Content

Keywords

Data science, machine learning, reinforcement learning, business analytics.

Learning Prerequisites

Required courses

Basic statistics and programming skills although are strongly encouraged for this course, ideally the student should have taken at least one course of each.  As we will do basic coding in Python in this course, some knowledge of Python is recommended.

Learning Outcomes

By the end of the course, the student must be able to:

• Formulate prediction and decision models
• Assess / Evaluate the performance of prediction and decision models
• Translate business objectives into measurable metrics, rewards, costs, and constraints
• Implement simple data science and reinforcement learning methods in Python
• Formulate simple business problems using states, actions, rewards, and transitions
• Assess / Evaluate decision policies using technical performance measures and business KPIs
• Infer information from data and critically analyse the results with respect to other method for a given business case
• Compare different business opportunities with respect to each other based on data

Transversal skills

• Access and evaluate appropriate sources of information.
• Take feedback (critique) and respond in an appropriate manner.
• Plan and carry out activities in a way which makes optimal use of available time and other resources.
• Assess one's own level of skill acquisition, and plan their on-going learning goals.
• Assess progress against the plan, and adapt the plan as appropriate.
• Collect data.

Teaching methods

Weekly lectures, demonstrations, assignments, and exercises.

Expected student activities

Attending class regularly to both acquire content and to review problem sets and exercises.  The teaching in the class is through use-cases, so attendance is highly recommended.

Assessment methods

20%     Individual assignment

40%     Group assignment and project

15%     Midterm Exam

25%     Final Exam

Supervision

Resources

Virtual desktop infrastructure (VDI)

No

Bibliography

There are many open-source materials online which we strongly encourage you to use. The following is the reading list for the course:

Data Science for Business: What You Need to Know about Data Mining and Data-Analytic Thinking by Foster Provost and Tom Fawcett. Published by O'Reilly Media. 1st edition (August 19, 2013) 414 pages ISBN-10: 1449361323

Reinforcement Learning: An Introduction by Richard S. Sutton and Andrew G. Barto. Published by The MIT Press. 2nd edition (November 13, 2018) 552 pages ISBN-10: 0262039249

Other material will appear during the course, so please stay attentive to this.

Ressources en bibliothèque

• Find the references at the Library

Moodle Link

• https://go.epfl.ch/MGT-432