Pattern Recognition

The content on this page was translated automatically.

Module namePattern Recognition
Type of moduleSelectable mandatory module
Learning results,
competencies, qualification goals
The student is able to:
  • explain the various tasks, models and algorithms of the pattern recognition,
  • create new modelling approaches to solve classification and regression problems,
  • plan and realize new applications independently,
  • critically question, compare and evaluate existing procedures and applications.

Learning results with regard to the objectives of the course of study:
  • Gaining deeper insight into the mathematical and natural science areas
  • Gaining a deeper knowledge about the specific electrical fundamentals
  • Acquiring enhanced and applied subject-specific basics
  • Identifying and classifying complex electro-technical and interdisciplinary tasks
  • Being confident in the ability to apply and evaluate analytical methods
  • Being able to create and evaluate solving methods independently
  • Familiarising oneself with new areas of knowledge, running searches and assessing the results
  • Gaining important and profound experience in the area of practical technical skills and engineering activities
  • Working and researching in national and international contexts
Types of courses4 SWS (semester periods per week):       3 SWS lecture
                                                                 1 SWS exercise
Course contentsThis lecture deals with the basic principles and procedures of the pattern recognition in particular from a probabilistic point of view. The following topics are discussed in the course: fundamentals (among other things stochastic, model selection, "Curse of Dimensionality", decision-making and information theory), distributions (multinomial, Dirichlet, Gaussian and student distribution, nonparametric estimation), linear models of regression, linear models of classification, mixed models and "Expectation Maximisation", approximate inference, combination of models, statistical learning theory (support vector machines), examples of applications (online clustering, anomaly detection, etc.).
Teaching and learning methods
(forms of teaching and learning)
Lecture, presentation, learning by teaching, self-regulated learning, problem-based learning
Frequency of the module offeringWinter term
Language German
Recommended (substantive) requirements for the participation in the moduleBasic knowledge about stochastic, analysis and linear algebra
Requirements for the
participation in the module
Prerequisites according to examination regulations
Student  workload180 h:   60 h attendance studies
                      120 h personal studies
Academic performancesWorking on exercises on a regular basis
Precondition for the
admission to the
examination performance
None
Examination performanceOral examination (20 min.)
Number of credits
of the module
6 credits

In charge of the moduleProf. Dr. Sick
Teacher of the moduleProf. Dr. Sick and co-workers
Forms of mediaPresentation with projector, paper exercises
Literature references
  • Lecture slides,
  • different sections of the book: Christopher M. Bishop: Pattern Recognition and Machine Learning, Springer (2006),
  • in addition, extracts from the book: Richard O. Duda, Peter E. Hart, David G. Stork: Pattern Classification, Wiley & Sons; 2nd edition (2000),
  • More reference literature is going to be recommended in the course.

                                                 Back to Modules page