Course detail

Classification and Recognition

FIT-IKRAcad. year: 2017/2018

The tasks of classification and pattern recognition, basic schema of a classifier, data and evaluation of individual methods, statistical pattern recognition, Bayes learning, maximum likelihood method, GMM, EM algorithm, discriminative training, kernel methods, hybrid systems, how to merge classifiers, basics of AdaBoost, structural recognition, speech processing applications - speaker recognition, language identification, speech recognition, keyword spotting, image processing - 2D object recognition, face detection, OCR, and natural language processing - document classification, text analysis.

Language of instruction

Czech

Number of ECTS credits

5

Mode of study

Not applicable.

Guarantor

doc. Ing. Lukáš Burget, Ph.D.

Department

Department of Computer Graphics and Multimedia (UPGM)

Learning outcomes of the course unit

The students will get acquainted with classification and recognition techniques and learn how to apply basic methods in the fields of speech recognition, computer graphics and natural language processing. They will understand the common aspects and differences of the particular methods and will be able to take advantage of the existing classifiers in real-situations.

The students will learn to work in a team. They will also improve their programming skills and their knowledge of development tools.

Prerequisites

Basic knowledge of the standard math notation.

Co-requisites

Not applicable.

Planned learning activities and teaching methods

Not applicable.

Assesment methods and criteria linked to learning outcomes

Study evaluation is based on marks obtained for specified items. Minimimum number of marks to pass is 50.

  • Realized project

Course curriculum

Syllabus of lectures:
  1. The tasks of classification and pattern recognition, basic schema of a classifier, data sets and evaluation
  2. Probabilistic distributions and linear models
  3. Statistical pattern recognition, Bayes learning, maximum likelihood method
  4. Sequential data modeling, hidden Markov models, linear dynamical systems
  5. Generative and discriminative models
  6. Speech processing applications - speaker recognition, language identification, speech recognition, keyword spotting
  7. Kernel methods
  8. Mixture models, EM algorithm
  9. Combining models, boosting
  10. AdaBoost, basics and extensions of the model
  11. Image processing - 2D object recognition, face detection, OCR
  12. Pattern recognition in text, grammars, languages, text analysis
  13. Project presentation, future directions

Syllabus - others, projects and individual work of students:
  • Individually assigned projects

Work placements

Not applicable.

Aims

To understand the foundations of classification and recognition and to learn how to apply basic algorithms and methods in this field to problems in speech recognition, computer graphics and natural language processing. To get acquainted with the evaluation procedures. To conceive basics of statistical pattern recognition, discriminative training and building hybrid systems.

Specification of controlled education, way of implementation and compensation for absences

The evaluation includes mid-term test, individual project, and the final exam. The mid-term test does not have a correction option, the final exam has two possible correction terms

Recommended optional programme components

Not applicable.

Prerequisites and corequisites

Basic literature

  • Bishop, C. M.: Pattern Recognition, Springer Science + Business Media, LLC, 2006, ISBN 0-387-31073-8.
  • Fukunaga, K. Statistical pattern recognition, Morgan Kaufmann, 1990, ISBN 0-122-69851-7.

Recommended reading

  • Mařík, V., Štěpánková, O., Lažanský, J. a kol.: Umělá inteligence (1-4), ACADEMIA Praha, 1998-2003, ISBN 80-200-1044-0.

Classification of course in study plans

  • Programme IT-BC-3 Bachelor's

    branch BIT , 2. year of study, summer semester, elective

Type of course unit

 

Lecture

26 hours, optionally

Teacher / Lecturer

doc. Ing. Lukáš Burget, Ph.D.

Syllabus

  1. The tasks of classification and pattern recognition, basic schema of a classifier, data sets and evaluation
  2. Probabilistic distributions and linear models
  3. Statistical pattern recognition, Bayes learning, maximum likelihood method
  4. Sequential data modeling, hidden Markov models, linear dynamical systems
  5. Generative and discriminative models
  6. Speech processing applications - speaker recognition, language identification, speech recognition, keyword spotting
  7. Kernel methods
  8. Mixture models, EM algorithm
  9. Combining models, boosting
  10. AdaBoost, basics and extensions of the model
  11. Image processing - 2D object recognition, face detection, OCR
  12. Pattern recognition in text, grammars, languages, text analysis
  13. Project presentation, future directions

Fundamentals seminar

13 hours, optionally

Teacher / Lecturer

doc. Ing. Lukáš Burget, Ph.D.

Project

13 hours, optionally

Teacher / Lecturer

doc. Ing. Lukáš Burget, Ph.D.