Machine Learning (4 cr)

Evaluation scale

0-5

Objective

You understand the basic principles of machine learning. You know the most common machine learning methods, you know how to apply them to existing data in practice, and how to interpret the results of the methods.

EUR-ACE Competences:
Knowledge and Understanding
Engineering Practice
Research and information retrieval

Content

- Supervised and unsupervised machine learning and the most common regression and classification models
- Application using Python libraries (NumPy, Pandas and scikit-learn)
- Data format and quality
- Splitting of the data set into training and test data
- Evaluation of model accuracy

Different models of machine learning:
- k-nearest neighbors
- k-means clustering
- Naive Bayes method
- Support Vector Machine
- Principal Component Analysis (PCA)
- Decision trees and random forest
- Perceptron (simple neural network)

Location and time

The course will be implemented in the spring semester of 2024.

Materials

The material for the assignments and the content to be studied will be shared during the course. The course utilizes the Python 3.9+ environment, Git version control, scikit-learn, Pandas, visualization libraries and other applicable libraries.

Teaching methods

Virtual study including doing assignments and familiarizing yourself with related lecture and example materials.

Employer connections

The aim is to connect the content of the course to problems that occur in working life.

Completion alternatives

The admission procedures are described in the degree rule and the study guide. The teacher of the course will give you more information on possible specific course practices.

Student workload

The workload of one credit corresponds to 27 hours of study. The total amount of study work (4 ECTS) in the course is 108 hours.

Assessment criteria, satisfactory (1)

Satisfactory 2: You know the most commonly used techniques of machine learning for various problems. You are able to choose the techniques of machine learning and apply your technical know-how in practice. In addition, you are able to assess your implementation superficially.

Sufficient 1: You know the most commonly used techniques in machine learning and are able to apply them. In addition, you are able to give a limited assessment of your implementation.

Assessment criteria, good (3)

Very good 4: You recognize the advantages of machine learning in the digital era. You know the most common techniques used in machine learning and are able to justify the use of the implemented techniques in various tasks. You are able to apply your technical know-how in practice and assess your implementation critically as well as validate its development.

Good 3: You are aware of the advantages of machine learning in the digital era. You knows the most commonly used techniques for various problems. You are able to apply your technical know-how in practice and validate its development.

Assessment criteria, excellent (5)

Excellent 5: You recognize the advantages of machine learning in the digital era. You master the techniques of machine learning in a versatile manner and are able to justify the use of implemented techniques in various tasks. You are able to apply your technical know-how in practice and assess your implementation critically as well as validate its development.

Qualifications

Basic ICT skills, basic skills in programming, knowledge and command of Python programming language.
Additionally, courses in Computational algorithms and Data Preprocessing .

Further information

The course is evaluated based on the returned assignments.
The assessment methods are reviewed at the beginning of the course.