Module CS4220 T

Module part: Pattern Recognition (MEa)


Duration

1 Semester

Turnus of offer

not available anymore

Credit points

4

Course of studies, specific fields and terms:

  • Master Computer Science 2019, module part, Module part
  • Master MES 2020, module part, computer science / electrical engineering
  • Master Entrepreneurship in Digital Technologies 2020, module part, Module part
  • Master IT-Security 2019, module part, Module part
  • Master Computer Science 2014, module part, advanced curriculum
  • Master Entrepreneurship in Digital Technologies 2014, module part, Module part
  • Master MES 2014, module part, computer science / electrical engineering
  • Master Computer Science 2014, Module part of a compulsory module, specialization field robotics and automation

Classes and lectures:

  • Pattern Recognition (lecture, 2 SWS)
  • Pattern Recognition (exercise, 1 SWS)

Workload:

  • 20 hours exam preparation
  • 45 hours in-classroom work
  • 55 hours private studies

Contents of teaching:

  • Introduction to probability theory
  • Principles of feature extraction and pattern recognition
  • Bayes decision theory
  • Discriminance functions
  • Neyman-Pearson test
  • Receiver Operating Characteristic
  • Parametric and nonparametric density estimation
  • kNN classifiers
  • Linear classifiers
  • Support vector machines and kernel trick
  • Random Forest
  • Neural Nets
  • Feature reduction and feature transforms
  • Validation of classifiers
  • Selected application scenarios: acoustic scene classification for the selection of hearing-aid algorithms, acoustic event recognition, attention classification based on EEG data, speaker and emotion recognition

Qualification-goals/Competencies:

  • Students are able to describe the main elements of feature extraction and pattern recognition.
  • They are able to explain the basic elements of statistical modeling.
  • They are able to use feature extraction, feature reduction and pattern classification techniques in practice.

Grading through:

  • exam type depends on main module

Responsible for this module:

Literature:

  • R. O. Duda, P. E. Hart, D. G. Storck : Pattern Classification New York: Wiley

Language:

  • offered only in German

Notes:

Admission requirements for the module:
- None

Admission requirements for the examination:
- Successful completion of the exercises during the semester (at least 50% of the achievable points).

Module Exam:
- CS4220-L1: Pattern Recognition, written exam, 90 min, 100% of module grade.

(Is equal to CS4220SJ14)
(Is module part of CS4510, CS4290, CS5274-KP08)

Last Updated:

28.08.2023