Course
DDML - Data Mining
PDF Course Catalog Deutsche Version: DDML
Version: 1 | Last Change: 12.02.2021 13:46 | Draft: 0 | Status: vom verantwortlichen Dozent freigegeben
| Long name | Data Mining |
|---|---|
| Approving CModule | DML_BaET |
| Responsible |
Prof. Dr. Beate Rhein
Professor Fakultät IME |
| Level | Bachelor |
| Semester in the year | winter semester |
| Duration | Semester |
| Hours in self-study | 78 |
| ECTS | 5 |
| Professors |
Prof. Dr. Beate Rhein
Professor Fakultät IME |
| Requirements | From Mathematics 1 and 2 the ability to construct mathematical models as well as knowledge of differential calculus and linear algebra is required. |
| Language | German |
| Separate final exam | Yes |
Literature
keine/none
Final exam
Details
Depending on the number of participants:For a small number of participants: combination of exam or oral examination and evaluation of the mini-project.
For many participants, examination by written examination; mini-project as prerequisite for participation in the examination.
In the written or oral examination, the methods, procedures, pitfalls and legal foundations of data mining are examined.
In the mini-project the ability to act independently and on one's own responsibility and the use of suitable software will be tested.
Minimum standard
Basic knowledge of the general approach to data mining, the procedures covered and their limitations.Exam Type
Depending on the number of participants:For a small number of participants: combination of exam or oral examination and evaluation of the mini-project.
For many participants, examination by written examination; mini-project as prerequisite for participation in the examination.
In the written or oral examination, the methods, procedures, pitfalls and legal foundations of data mining are examined.
In the mini-project the ability to act independently and on one's own responsibility and the use of suitable software will be tested.
Learning goals
Knowledge
Introduction to a suitable software, e.g. Python
Introduction to descriptive statistics and possibly also probability calculation
Supervised learning:
- Classification procedure: Procedure, performance measures, application of a method of instance-based learning, e.g. k-nearest-neighbor and a method of model-based learning, e.g. decision trees
- Possibly regression analysis: about machine learning and classical
Unsupervised learning:
- Cluster analysis: k-means, possibly also DBSCAN
Preprocessing of the data:
- Handling Damaged / Missing Data
- Runaway or noise - problems
- Scaling
- Visualization of data
- Possible dimension reduction
- Assessment of data quality
- possibly look at different types of data records, make reference to NoSql databases, Outlook on current research, e.g. image recognition, Natural Language Processing, Reinforcement Learning
Introduction to descriptive statistics and possibly also probability calculation
Supervised learning:
- Classification procedure: Procedure, performance measures, application of a method of instance-based learning, e.g. k-nearest-neighbor and a method of model-based learning, e.g. decision trees
- Possibly regression analysis: about machine learning and classical
Unsupervised learning:
- Cluster analysis: k-means, possibly also DBSCAN
Preprocessing of the data:
- Handling Damaged / Missing Data
- Runaway or noise - problems
- Scaling
- Visualization of data
- Possible dimension reduction
- Assessment of data quality
- possibly look at different types of data records, make reference to NoSql databases, Outlook on current research, e.g. image recognition, Natural Language Processing, Reinforcement Learning
Skills
Be able to name and apply a suitable method and overall approach to tasks
Select and evaluate a suitable performance measure
Apply Privacy Policy
Select and evaluate a suitable performance measure
Apply Privacy Policy
Expenditure classroom teaching
| Type | Attendance (h/Wk.) |
|---|---|
| Lecture | 2 |
| Exercises (whole course) | 0 |
| Exercises (shared course) | 2 |
| Tutorial (voluntary) | 0 |
Special literature
A. Geron: Praxiseinstieg Machine Learning mit Scikit-Learn und TensorFlow: Konzepte, Tools und Techniken für intelligente Systeme, Heidelberg, o‘Reilly Verlag 2017, 978-3960090618, S. Raschka, V. Mirjalili: Machine Learning mit Python und Scikit-Learn und TensorFlow: Das umfassende Praxis-Handbuch für Data Science, Predictive Analytics und Deep Learning, mitp Verlag, 2018, 978-3958457331, J. Frochte, Jörg: Maschinelles Lernen, München, Carl Hanser Verlag GmbH & Co. KG, 2018, eBook ISBN: 978-3-446-45705-8, Print ISBN: 978-3-446-45291-6, A. Müller: Einführung in Machine Learning mit Python: Praxiswissen Data Science, Heidelberg, o‘Reilly Verlag 2017, eBook: 978-3-96010-111-6
Special requirements
none
Accompanying material
Script or set of slides
Tasks (expected to be integrated into the script)
Mini project task with data set
Tasks (expected to be integrated into the script)
Mini project task with data set
Separate exam
none
Bei Fehlern, bitte Mitteilung an die
Webredaktion der Fakultät IME
Webredaktion der Fakultät IME
© 2022 Technische Hochschule Köln