Syllabus

Title
6019 Exploiting Data: The Machine Learning Approach
Instructors
Mgr. Jan Valendin, Stefan Vamosi, MSc.
Contact details
  • Type
    PI
  • Weekly hours
    2
  • Language of instruction
    Englisch
Registration
02/21/19 to 02/27/19
Registration via LPIS
Notes to the course
Subject(s) Bachelor Programs
Dates
Day Date Time Room
Tuesday 03/19/19 06:30 PM - 09:30 PM TC.-1.61
Tuesday 03/26/19 06:30 PM - 09:30 PM TC.-1.61
Tuesday 04/02/19 06:30 PM - 09:30 PM TC.-1.61
Tuesday 04/09/19 06:30 PM - 09:30 PM TC.-1.61
Tuesday 05/07/19 06:30 PM - 08:45 PM LC.-1.038
Tuesday 05/21/19 06:30 PM - 09:30 PM D2.-1.019 Workstation-Raum
Tuesday 05/28/19 06:30 PM - 09:30 PM D2.-1.019 Workstation-Raum
Tuesday 06/04/19 01:00 PM - 03:30 PM TC.4.02

Contents

The fast rise of available data for customer and market analysis creates an immense opportunity for machine learning applications. For predictive tasks, these models often outperform conventional statistical approaches. This course introduces some of the state-of-the-art machine learning techniques including the basic theoretical background with hands-on exercises and group works.

The course includes a short introduction to R, supervised machine learning, model performance evaluation, cross-validation and generalization. We will focus on the application and the concept of the following machine learning methods: k-nearest neighbors (kNN), decision trees, random forests, support vector machines (SVM), deep learning and k-means clustering.

Learning outcomes

The learning objectives of this course are as follows:

  • Understand the basic theoretical background of various supervised machine learning algorithms.
  • Get familiar with various machine learning algorithms in R by exercising on real data examples.
  • Gain the knowledge to pick the right machine learning method for a particular problem.

Attendance requirements

You need to attend at least 80% of all classes to pass the course.

Teaching/learning method(s)

This course uses various teaching methods like classic lecture format, guided computer exercises, group works and short student presentations.

Assessment

The final grade will be evaluated as follows:

  • Class participation: [weight: 10%]
  • Online Assessment for R (Data Camp): [weight: 10%]
  • Individual computer exercises (4 homeworks): [weight: 20%]
  • Final exam (open questions and multiple choice): [weight: 50%]
  • Completion of the group challenge: [weight: 10%]
  • Winners of the group challenge: [weight: 5%] Bonus

For the individual home exercises, a random draw will select students to make a short presentation about their findings. If a presentation has failed (no preparation) the final grade will be reduced by one grade.

To successfully pass this course, your weighted final grade needs to exceed 60%.

Readings

1 Author: Brett Lantz
Title:

Machine Learning with R


Publisher: Packt Publishing
Edition: 2
Year: 2015
Content relevant for class examination: Yes
Content relevant for diploma examination: No
Recommendation: Reference literature
Type: Book
2 Author: Ian Goodfellow, Yoshua Bengio, Aaron Courville
Title:

Deep Learning


Publisher: mitp
Remarks: www.deeplearningbook.org/
Recommendation: Reference literature
Type: Book

Recommended previous knowledge and skills

Basic knowledge of standard statistical software packages, such as R (or Python) are recommended. For those who are not familiar with R and to ensure a shared set of R skills among all participants, we will provide a condensed introduction to R on the first day and an online exercise.

Last edited: 2019-01-24



Back