After successful completion of the course, students are able to design, realize, and evaluate simple pattern recognition systems, in particular regression, binary and multiclass classification, and clustering.

• Design of pattern recognition systems
• Features: extraction, selection, reduction, PCA
• Regression and classification, k-Nearest Neighbors
• Bayes theorem
• Linear discriminants, Perceptron
• Neural networks
• Decision trees
• Markov models
• Evaluation of pattern recognition systems
• Clustering

• Assignments are completed individualy
• Assignments are solved in Python using Jupyter notebooks
• Lecture slides and additional documentation will be provided
• TUWEL is used to distribute materials and to coordinate the administration of this UE and the associated VO

Students can achieve a total of 100 points from 4 assignments. The grading is based on the toal amount of points.