Machine Learning with R, part 1, 7.5 Credits

Contents

This course provides a gentle introduction to machine learning. The basic ideas and core concepts of machine learning will be introduced. Building on this, it will focus on linear models and their role in machine learning. More specifically, the course will mainly discuss Gaussian discriminant analysis and logistic regression model for classification problems and multiple regression model for continuous variable prediction problems. Ultimately, students will develop an awareness of using data and models to solve problems. In addition, as a platform for the implementation of different approaches, the R programming language and its environment will be introduced. The implementation of the various methods in the R environment will also be shown.

Expected learning outcomes

Knowledge and understanding

Students must be able to
1.    Explain the basic ideas and objectives of machine learning.
2.    Explain the mechanism of operation and characteristics of the linear models for machine learning that are covered in the course.
3.    Explain the problem of overfitting in machine learning

Skills and ability

Students must be able to
4.    Transfer concrete problems into machine learning problems and select appropriate methods and models to solve them.
5.    Use R to implement methods for training, model selection and model validation that are introduced in the course.
6.    Find and access documentation and help pages for R to gain knowledge about available R-packages and to use them.

Judgement and approach

Students must be able to
7.    Choose, based on appropriate criteria for the problem at hand, the best model among candidates using different validation methods.

Required Knowledge

7,5 ECTS in Statistics incl. Regression Analysis. Proficiency in English equivalent to Swedish upper secondary course English B/6

Form of instruction

Learning is supported by lectures, lessons, computer lab sessions, and tutoring.

Examination modes

The examination consists of two parts: a written home exam using R and a following oral exam. The grades used are VG (Pass with distinction), G (Pass), and U (Fail). The grade awarded on the course is based on an overall assessment of the two parts.

A student who has passed an examination is not allowed to take another examination in order to get a higher grade. For students who do not pass, an additional test will be held according to a set schedule.

Exceptions from examination form as stated in the syllabus can be made for a student who has a decision on pedagogical support for disabilities. Individual adaptations of the examination form should be considered based on the student's needs. The examination form shall be adapted within the framework of the expected learning outcomes stated in the course syllabus. At the request of the student, the course's responsible teacher, in consultation with the examiner, must promptly decide on the adapted examination form.  The student must then be informed of the decision.

After two failed examinations in a course, or part of a course, the student has the right to request another examiner unless there are special reasons against it. The request for a new examiner is made to the Director of studies at the Department of Statistics.

Examinations based on the same course syllabus as the ordinary examinations are guaranteed to be offered up to two years after the date of the student's first registration for the course.

Academic credit transfer
Academic credit transfers are according to the University credit transfer regulations.

Other regulations

his course partly overlaps with the course Introduction to data science (3 Credits) and the course Statistics C: Methods for Data Science (10 Credits) and cannot therefore be credited in the degree together with any of the overlapping courses.