Syllabus

Title
1909 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
09/19/19 to 09/26/19
Registration via LPIS
Notes to the course
Subject(s) Bachelor Programs
Dates
Day Date Time Room
Tuesday 10/15/19 05:00 PM - 08:00 PM TC.4.02
Tuesday 10/22/19 05:00 PM - 08:00 PM TC.4.02
Tuesday 10/29/19 05:00 PM - 08:00 PM TC.4.02
Tuesday 11/05/19 05:00 PM - 08:00 PM TC.4.02
Tuesday 11/12/19 05:00 PM - 08:00 PM TC.4.02
Tuesday 11/19/19 05:00 PM - 08:00 PM TC.4.02
Tuesday 12/03/19 05:00 PM - 08:00 PM TC.4.02
Tuesday 12/10/19 05:00 PM - 07:00 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 modern 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, k-means clustering, neural networks and support vector machines (SVM).

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%]
  • Individual computer exercises (5 homeworks): [weight: 50%]
  • Final exam (open questions and multiple choice): [weight: 40%]
  • Group challenge: [weight: 10% Bonus in Total]

For the individual home exercises, a random draw will select students to make a short presentation about their findings. If a presentation has failed, the final grade will be reduced by one grade.

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

Prerequisites for participation and waiting lists

The class will not take place in a computer room. Please bring your own Notebook with an R-Studio installation to attend the class. In exceptional cases, a Notebook can be provided by the institute.

The software is open source and can be downloaded from: www.rstudio.com

Readings

1 Author: Brett Lantz
Title:

Machine Learning with R


Publisher: Packt Publishing
Edition: 2
Year: 2015
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: 2019-08-26



Back