Module MA4665-KP05
Statistical Learning (StaLerKP05)
Duration
1 Semester
Turnus of offer
every second year
Credit points
5
Course of studies, specific fields and terms:
- Master Medical Informatics 2019, optional subject, Medical Data Science / Artificial Intelligence
- Master CLS 2023, optional subject, mathematics
- Bachelor CLS 2023, optional subject, mathematics
- Master CLS 2016, optional subject, mathematics
- Bachelor CLS 2016, optional subject, mathematics
Classes and lectures:
- Statistical Learning (exercise, 1 SWS)
- Statistical Learning (lecture, 2 SWS)
Workload:
- 45 hours in-classroom work
- 15 hours exam preparation
- 60 hours private studies
- 30 hours work on project
Contents of teaching:
- Application scenarios and research questions for prediction models (focus: risk prediction)
- Study design and data preprocessing
- Overview of different machine learning methods (concepts, advantages and disadvantages)
- Development of prediction models
- Evaluation of prediction performance
- Comparison of prediction models
- Variable selection
- Extension to time-to-event outcomes with censoring
Qualification-goals/Competencies:
- Students can define research questions for applications of pediction models
- They can explain the individual steps in the development and evaluation of prediction models
- They can describe frequently occurring errors and problems as well als possible solutions
- They can describe central ideas of different machine learning methods and select suitable methods for applications
- They can develop and evaluate models in the programming language R
Grading through:
- project work
- Viva Voce or test
Responsible for this module:
- Prof. Dr. rer. nat. Silke Szymczak
Teacher:
- Institute of Medical Biometry and Statistics
- Prof. Dr. rer. nat. Silke Szymczak
- MitarbeiterInnen des Instituts
Literature:
- Thomas Gerds und Michael Kattan : Medical Risk Prediction Models With Ties to Machine Learning CRC Press: Bota Raton, FL (2022)
Language:
- German or English
Notes:
Admission requirements for taking the module:- None (The competencies of the modules listed under 'Requires' are needed for this module, but are not a formal prerequisite)
Admission requirements for participation in module examination(s):
- None
Module exam(s):
- MA4665-L1: Statistical Learning, oral exam (20 min) or written exam (60 min), 50 % of module grade
- MA4665-L2: Research project incl. presentation and code documentation, 50 % of module grade
Last Updated:
12.09.2024