BAN442 From data to value: Machine Learning with Python

• Topics

  This course introduces the student to machine learning (ML) in Python using practical examples. The focus will be on applying ML to a business case and create value from data.

  As businesses continue to evolve towards becoming more data driven, we need tools to take advantage of the data. One of these tools is machine learning, which can process and make use of vast amounts of data that are not comprehensible for humans. Hence, ML can be utilized to improve decision support by using data that is not fully exploited yet. This course is an introduction to data driven decision processes and how they are used in business applications.

  While machine learning can be done in several programming languages, Python continues to be a top choice for many professionals and first-time coders. The simplicity and widespread use of Python makes it a very good choice for business students as it can be used for building ML-systems, automating tasks and develop applications.

  Topics covered in this course:

  • Basics of Python
  • How can ml be an appropriate tool for use in decision support
  • Assess the quality of data
  • Solving a business case using ml and Python

  The course is intended for students with little to no prior knowledge of python, but with some knowledge of statistical/machine learning.

• Learning outcome

  Upon completion of the course, students have been familiarized with Python, machine learning, the value ML can create for business stakeholders and have:

  the practical skills to

  • Solve a business problem using machine learning and Python
  • Set up an environment for programming in Python
  • Prepare data for ML
  • Use Python as a tool for creating ML-solutions
  • Create simple programs using Python

  and the knowledge to

  • Understand the basics of Python
  • Address problems relating to data quality
  • Address over-/under fitting
  • Decide when different ML-methods are appropriate to use
  • Understand the concept of data drift and the implications for ML-solutions

• Teaching

  This is a one-week intensive course that consists of daily lectures and in-class assignments to be solved in groups.

• Recommended prerequisites

  The course introduces the students to applied machine learning with Python. Hence, a prior knowledge in machine learning from BAN404 is advised. Basic skills in Python programming are helpful, but not required to follow the course.

• Credit reduction due to overlap

  None.

• Compulsory Activity

  Approved homework assignments.

• Assessment

  Assigned, group term paper (2-3 students in each group) in which the group will demonstrate the tools and concepts learned in the course.

  The group assignment will be given at the end of the course, and the students will have two weeks (weeks 35 and 36) to complete the assignment.

• Grading Scale

  Pass-Fail.

• Computer tools

  Python and Jupyter Lab. Students are recommended to download the Anaconda distribution for Python. More details regarding the required software will be provided at the beginning of the course.

• Literature

  To be announced in Canvas.