Course Information

Bi0034 Analysis and classification of data

Extent and Intensity

2/1/0. 3 credit(s) (plus extra credits for completion). Recommended Type of Completion: zk (examination). Other types of completion: z (credit).

Teacher(s)

RNDr. Eva Koriťáková, Ph.D. (lecturer)
prof. Ing. Jiří Holčík, CSc. (lecturer)
RNDr. Roman Vyškovský, Ph.D. (lecturer)
Mgr. Tereza Jurková (seminar tutor)

Guaranteed by

prof. Ing. Jiří Holčík, CSc.
RECETOX – Faculty of Science
Contact Person: RNDr. Eva Koriťáková, Ph.D.
Supplier department: RECETOX – Faculty of Science

Timetable

Wed 8:00–10:50 D29/347-RCX2

Prerequisites

It is recommended to attend the course Bi8600 Multivariate Methods or the course Bi8601 Advanced Statistical Methods beforehand.

Course Enrolment Limitations

The course is also offered to the students of the fields other than those the course is directly associated with.

fields of study / plans the course is directly associated with

• Biomedical bioinformatics (programme PřF, N-MBB)
• Epidemiology and modeling (programme PřF, N-MBB)

Course objectives

The aim of the course is to introduce objectives and principles of data analysis and classification methods to students. During the course, the students will acquire knowledge of fundamental theoretical and methodological principles of pattern analysis and recognition methods with emphasis to processing of medical and biological data. Besides, they will master application of the methods on real-world data and interpretation of achieved results.

Learning outcomes

At the end of the course, students should be able to:
- Choose and apply appropriate data analysis method to obtain required results;
- Reduce and transform multivariate data using advanced ordination methods;
- Classify data using various classification methods;
- Design modified algorithms to process data of given particular characteristics;
- Interpret obtained results including evaluation of classification performance.

Syllabus

• Introduction to analysis and classification of data.
• Classification based on discriminant functions.
• Classification based on minimal distance – deepening of knowledge about distance and similarity metrics.
• Classification based on minimal distance.
• Classification based on boundaries – Fisher’s Linear Discriminant Analysis (FLDA).
• Classification based on boundaries – Support Vector Machines (SVM).
• Sequential classification.
• Evaluation of classification performance – cross-validation, comparison of classification results with random classification, comparison of classification performance of two or more classifiers.
• Feature extraction – deepening of knowledge about Principal Component Analysis (PCA).
• Advanced methods for feature extraction – Manifold Learning, Independent Component Analysis (ICA).
• Feature selection.
• Application of classification, pattern analysis and recognition methods on real-world data.

Literature

• Mitchel,T.M.: Machine Learning. McGraw Hill 1997
• Dunham,M.H.: Data Mining: Introductory and Advanced Topics. Prentice Hall 2002
• Holčík,J.: Analýza a klasifikace signálů. [Učební texty vysokých škol] Brno, Nakladatelství VUT v Brně 1992.

Teaching methods

Online lectures and full-time form of practices. Understanding of principles, methods and algorithms is emphasized.

Assessment methods

oral examination

Language of instruction

Czech

Further Comments

Study Materials
The course is taught annually.

Bi0034 Analysis and classification of data

Extent and Intensity

2/1/0. 3 credit(s) (plus extra credits for completion). Recommended Type of Completion: zk (examination). Other types of completion: z (credit).

Teacher(s)

RNDr. Eva Koriťáková, Ph.D. (lecturer)
prof. Ing. Jiří Holčík, CSc. (lecturer)
RNDr. Roman Vyškovský, Ph.D. (lecturer)

Guaranteed by

prof. Ing. Jiří Holčík, CSc.
RECETOX – Faculty of Science
Contact Person: RNDr. Eva Koriťáková, Ph.D.
Supplier department: RECETOX – Faculty of Science

Timetable

Wed 8:00–10:50 D29/347-RCX2

Prerequisites

It is recommended to attend the course Bi8600 Multivariate Methods or the course Bi8601 Advanced Statistical Methods beforehand.

Course Enrolment Limitations

The course is also offered to the students of the fields other than those the course is directly associated with.

fields of study / plans the course is directly associated with

• Biomedical bioinformatics (programme PřF, N-MBB)
• Epidemiology and modeling (programme PřF, N-MBB)

Course objectives

The aim of the course is to introduce objectives and principles of data analysis and classification methods to students. During the course, the students will acquire knowledge of fundamental theoretical and methodological principles of pattern analysis and recognition methods with emphasis to processing of medical and biological data. Besides, they will master application of the methods on real-world data and interpretation of achieved results.

Learning outcomes

At the end of the course, students should be able to:
- Choose and apply appropriate data analysis method to obtain required results;
- Reduce and transform multivariate data using advanced ordination methods;
- Classify data using various classification methods;
- Design modified algorithms to process data of given particular characteristics;
- Interpret obtained results including evaluation of classification performance.

Syllabus

• Introduction to analysis and classification of data.
• Classification based on discriminant functions.
• Classification based on minimal distance – deepening of knowledge about distance and similarity metrics.
• Classification based on minimal distance.
• Classification based on boundaries – Fisher’s Linear Discriminant Analysis (FLDA).
• Classification based on boundaries – Support Vector Machines (SVM).
• Sequential classification.
• Evaluation of classification performance – cross-validation, comparison of classification results with random classification, comparison of classification performance of two or more classifiers.
• Feature extraction – deepening of knowledge about Principal Component Analysis (PCA).
• Advanced methods for feature extraction – Manifold Learning, Independent Component Analysis (ICA).
• Feature selection.
• Application of classification, pattern analysis and recognition methods on real-world data.

Literature

• Mitchel,T.M.: Machine Learning. McGraw Hill 1997
• Dunham,M.H.: Data Mining: Introductory and Advanced Topics. Prentice Hall 2002
• Holčík,J.: Analýza a klasifikace signálů. [Učební texty vysokých škol] Brno, Nakladatelství VUT v Brně 1992.

Teaching methods

Online lectures using Microsoft Teams supported by Power Point presentations. Understanding of principles, methods and algorithms is emphasized. Students are continuously encouraged to interact with the lecturer.

Assessment methods

oral examination

Language of instruction

Czech

Further Comments

Study Materials
The course is taught annually.

Bi0034 Analysis and classification of data

Extent and Intensity

2/1/0. 3 credit(s) (plus extra credits for completion). Recommended Type of Completion: zk (examination). Other types of completion: z (credit).

Teacher(s)

RNDr. Eva Koriťáková, Ph.D. (lecturer)
prof. Ing. Jiří Holčík, CSc. (lecturer)
RNDr. Roman Vyškovský, Ph.D. (lecturer)

Guaranteed by

prof. Ing. Jiří Holčík, CSc.
RECETOX – Faculty of Science
Contact Person: RNDr. Eva Koriťáková, Ph.D.
Supplier department: RECETOX – Faculty of Science

Timetable

Wed 8:00–10:50 D29/347-RCX2

Prerequisites

It is recommended to attend the course Bi8600 Multivariate Methods or the course Bi8601 Advanced Statistical Methods beforehand.

Course Enrolment Limitations

The course is also offered to the students of the fields other than those the course is directly associated with.

fields of study / plans the course is directly associated with

• Biomedical bioinformatics (programme PřF, N-MBB)
• Epidemiology and modeling (programme PřF, N-MBB)

Course objectives

The aim of the course is to introduce objectives and principles of data analysis and classification methods to students. During the course, the students will acquire knowledge of fundamental theoretical and methodological principles of pattern analysis and recognition methods with emphasis to processing of medical and biological data. Besides, they will master application of the methods on real-world data and interpretation of achieved results.

Learning outcomes

At the end of the course, students should be able to:
- Choose and apply appropriate data analysis method to obtain required results;
- Reduce and transform multivariate data using advanced ordination methods;
- Classify data using various classification methods;
- Design modified algorithms to process data of given particular characteristics;
- Interpret obtained results including evaluation of classification performance.

Syllabus

• Introduction to analysis and classification of data.
• Classification based on discriminant functions.
• Classification based on minimal distance – deepening of knowledge about distance and similarity metrics.
• Classification based on minimal distance.
• Classification based on boundaries – Fisher’s Linear Discriminant Analysis (FLDA).
• Classification based on boundaries – Support Vector Machines (SVM).
• Sequential classification.
• Evaluation of classification performance – cross-validation, comparison of classification results with random classification, comparison of classification performance of two or more classifiers.
• Feature extraction – deepening of knowledge about Principal Component Analysis (PCA).
• Advanced methods for feature extraction – Manifold Learning, Independent Component Analysis (ICA).
• Feature selection.
• Application of classification, pattern analysis and recognition methods on real-world data.

Literature

• Mitchel,T.M.: Machine Learning. McGraw Hill 1997
• Dunham,M.H.: Data Mining: Introductory and Advanced Topics. Prentice Hall 2002
• Holčík,J.: Analýza a klasifikace signálů. [Učební texty vysokých škol] Brno, Nakladatelství VUT v Brně 1992.

Teaching methods

Lectures supported by Power Point presentations. Understanding of principles, methods and algorithms is emphasized. Students are continuously encouraged to interact with the lecturer.

Assessment methods

oral examination

Language of instruction

Czech

Further Comments

Study Materials
The course is taught annually.

Bi0034 Analysis and classification of data

Extent and Intensity

2/1/0. 3 credit(s) (plus extra credits for completion). Recommended Type of Completion: zk (examination). Other types of completion: z (credit).

Teacher(s)

RNDr. Eva Koriťáková, Ph.D. (lecturer)
prof. Ing. Jiří Holčík, CSc. (lecturer)

Guaranteed by

prof. RNDr. Ladislav Dušek, Ph.D.
RECETOX – Faculty of Science
Contact Person: RNDr. Eva Koriťáková, Ph.D.
Supplier department: RECETOX – Faculty of Science

Timetable

Mon 17. 9. to Fri 14. 12. Wed 8:00–10:50 D29/347-RCX2

Prerequisites

It is recommended to attend the course Bi8600 Multivariate Methods beforehand.

Course Enrolment Limitations

The course is offered to students of any study field.

Course objectives

At the end of the course, students should be able to: - know fundamental theoretical and methodological principles of pattern analysis and recognition with emphasis to processing of medical and biological data;- choose and apply appropriatedata analysis method to obtain required results; - reduce and transform multivariate data using advances ordination methods; - classify data using various classification methods; - design modified algorithms to process data of given particular characteristics; - interpret obtained results including evaluation of classification performance.

Syllabus

• 1. Multivariate data analysis and its aims. 2. Ordination methods – factor analysis (FA). 3. Ordination methods – canonical correspondence analysis (CCA), redundancy analysis (RDA). 4. Ordination methods – co-coordinate analysis. 5. Advanced method of feature extraction – independent component analysis (ICA) and other advanced methods. 6. Distance and similarity metrics. 7. Classification – introduction, discriminant analysis based on discriminant functions and minimal distance. 8. Classification – discriminant analysis based on boundaries: Fisher’s linear discriminant analysis. 9. Classification – discriminant analysis based on boundaries: linear support vector machines (SVM). 10. Classification – nonlinearSVM, sequential classification. 11. Evaluation of classification performance – cross-validation, comparison of classification results with random classification, comparison of classification performance of two or more classifiers. 12. Summary of methods for data analysis and classification.

Literature

• Dunham,M.H.: Data Mining: Introductory and Advanced Topics. Prentice Hall 2002
• Holčík,J.: Analýza a klasifikace signálů. [Učební texty vysokých škol] Brno, Nakladatelství VUT v Brně 1992.
• Mitchel,T.M.: Machine Learning. McGraw Hill 1997

Teaching methods

Lectures supported by Power Point presentations. Understanding of principles, methods and algorithms is emphasized. Students are continuously encouraged to interact with the lecturer.

Assessment methods

oral examination

Language of instruction

Czech

Further Comments

Study Materials
The course is taught annually.

Bi0034 Analysis and classification of data

Extent and Intensity

2/1/0. 3 credit(s) (plus extra credits for completion). Recommended Type of Completion: zk (examination). Other types of completion: z (credit).

Teacher(s)

RNDr. Eva Koriťáková, Ph.D. (lecturer)
prof. Ing. Jiří Holčík, CSc. (lecturer)

Guaranteed by

prof. RNDr. Ladislav Dušek, Ph.D.
RECETOX – Faculty of Science
Contact Person: RNDr. Eva Koriťáková, Ph.D.
Supplier department: RECETOX – Faculty of Science

Timetable

Mon 18. 9. to Fri 15. 12. Wed 8:00–10:50 D29/347-RCX2

Prerequisites

It is recommended to attend the course Bi8600 Multivariate Methods beforehand.

Course Enrolment Limitations

The course is offered to students of any study field.

Course objectives

At the end of the course, students should be able to: - know fundamental theoretical and methodological principles of pattern analysis and recognition with emphasis to processing of medical and biological data;- choose and apply appropriatedata analysis method to obtain required results; - reduce and transform multivariate data using advances ordination methods; - classify data using various classification methods; - design modified algorithms to process data of given particular characteristics; - interpret obtained results including evaluation of classification performance.

Syllabus

• 1. Multivariate data analysis and its aims. 2. Ordination methods – factor analysis (FA). 3. Ordination methods – canonical correspondence analysis (CCA), redundancy analysis (RDA). 4. Ordination methods – co-coordinate analysis. 5. Advanced method of feature extraction – independent component analysis (ICA) and other advanced methods. 6. Distance and similarity metrics. 7. Classification – introduction, discriminant analysis based on discriminant functions and minimal distance. 8. Classification – discriminant analysis based on boundaries: Fisher’s linear discriminant analysis. 9. Classification – discriminant analysis based on boundaries: linear support vector machines (SVM). 10. Classification – nonlinearSVM, sequential classification. 11. Evaluation of classification performance – cross-validation, comparison of classification results with random classification, comparison of classification performance of two or more classifiers. 12. Summary of methods for data analysis and classification.

Literature

• Dunham,M.H.: Data Mining: Introductory and Advanced Topics. Prentice Hall 2002
• Holčík,J.: Analýza a klasifikace signálů. [Učební texty vysokých škol] Brno, Nakladatelství VUT v Brně 1992.
• Mitchel,T.M.: Machine Learning. McGraw Hill 1997

Teaching methods

Lectures supported by Power Point presentations. Understanding of principles, methods and algorithms is emphasized. Students are continuously encouraged to interact with the lecturer.

Assessment methods

oral examination

Language of instruction

Czech

Further Comments

Study Materials
The course is taught annually.

Bi0034 Analysis and classification of data

Extent and Intensity

2/1/0. 3 credit(s) (plus extra credits for completion). Recommended Type of Completion: zk (examination). Other types of completion: z (credit).

Teacher(s)

RNDr. Eva Koriťáková, Ph.D. (lecturer)
prof. Ing. Jiří Holčík, CSc. (lecturer)

Guaranteed by

prof. RNDr. Ladislav Dušek, Ph.D.
RECETOX – Faculty of Science
Contact Person: RNDr. Eva Koriťáková, Ph.D.
Supplier department: RECETOX – Faculty of Science

Timetable

Mon 19. 9. to Sun 18. 12. Wed 8:00–10:50 D29/347-RCX2

Prerequisites

It is recommended to attend the course Bi8600 Multivariate Methods beforehand.

Course Enrolment Limitations

The course is offered to students of any study field.

Course objectives

At the end of the course, students should be able to: - know fundamental theoretical and methodological principles of pattern analysis and recognition with emphasis to processing of medical and biological data;- choose and apply appropriatedata analysis method to obtain required results; - reduce and transform multivariate data using advances ordination methods; - classify data using various classification methods; - design modified algorithms to process data of given particular characteristics; - interpret obtained results including evaluation of classification performance.

Syllabus

• 1. Multivariate data analysis and its aims. 2. Ordination methods – factor analysis (FA). 3. Ordination methods – canonical correspondence analysis (CCA), redundancy analysis (RDA). 4. Ordination methods – co-coordinate analysis. 5. Advanced method of feature extraction – independent component analysis (ICA) and other advanced methods. 6. Distance and similarity metrics. 7. Classification – introduction, discriminant analysis based on discriminant functions and minimal distance. 8. Classification – discriminant analysis based on boundaries: Fisher’s linear discriminant analysis. 9. Classification – discriminant analysis based on boundaries: linear support vector machines (SVM). 10. Classification – nonlinearSVM, sequential classification. 11. Evaluation of classification performance – cross-validation, comparison of classification results with random classification, comparison of classification performance of two or more classifiers. 12. Summary of methods for data analysis and classification.

Literature

• Dunham,M.H.: Data Mining: Introductory and Advanced Topics. Prentice Hall 2002
• Holčík,J.: Analýza a klasifikace signálů. [Učební texty vysokých škol] Brno, Nakladatelství VUT v Brně 1992.
• Mitchel,T.M.: Machine Learning. McGraw Hill 1997

Teaching methods

Lectures supported by Power Point presentations. Understanding of principles, methods and algorithms is emphasized. Students are continuously encouraged to interact with the lecturer.

Assessment methods

oral examination

Language of instruction

Czech

Further Comments

Study Materials
The course is taught annually.

Bi0034 Analysis and classification of data

Extent and Intensity

2/1/0. 3 credit(s) (plus extra credits for completion). Recommended Type of Completion: zk (examination). Other types of completion: z (credit).

Teacher(s)

prof. Ing. Jiří Holčík, CSc. (lecturer)
RNDr. Eva Koriťáková, Ph.D. (lecturer)

Guaranteed by

prof. RNDr. Ladislav Dušek, Ph.D.
RECETOX – Faculty of Science
Contact Person: prof. Ing. Jiří Holčík, CSc.
Supplier department: RECETOX – Faculty of Science

Timetable

Wed 8:00–10:50 D29/347-RCX2

Prerequisites

It is recommended to attend the course Bi8600 Multivariate Methods beforehand.

Course Enrolment Limitations

The course is offered to students of any study field.

Course objectives

At the end of the course, students should be able to: - know fundamental theoretical and methodological principles of pattern analysis and recognition with emphasis to processing of medical and biological data;- choose and apply appropriatedata analysis method to obtain required results; - reduce and transform multivariate data using advances ordination methods; - classify data using various classification methods; - design modified algorithms to process data of given particular characteristics; - interpret obtained results including evaluation of classification performance.

Syllabus

• 1. Multivariate data analysis and its aims. 2. Ordination methods – factor analysis (FA). 3. Ordination methods – canonical correspondence analysis (CCA), redundancy analysis (RDA). 4. Ordination methods – co-coordinate analysis. 5. Advanced method of feature extraction – independent component analysis (ICA) and other advanced methods. 6. Distance and similarity metrics. 7. Classification – introduction, discriminant analysis based on discriminant functions and minimal distance. 8. Classification – discriminant analysis based on boundaries: Fisher’s linear discriminant analysis. 9. Classification – discriminant analysis based on boundaries: linear support vector machines (SVM). 10. Classification – nonlinearSVM, sequential classification. 11. Evaluation of classification performance – cross-validation, comparison of classification results with random classification, comparison of classification performance of two or more classifiers. 12. Summary of methods for data analysis and classification.

Literature

• Dunham,M.H.: Data Mining: Introductory and Advanced Topics. Prentice Hall 2002
• Holčík,J.: Analýza a klasifikace signálů. [Učební texty vysokých škol] Brno, Nakladatelství VUT v Brně 1992.
• Mitchel,T.M.: Machine Learning. McGraw Hill 1997

Teaching methods

Lectures supported by Power Point presentations. Understanding of principles, methods and algorithms is emphasized. Students are continuously encouraged to interact with the lecturer.

Assessment methods

oral examination

Language of instruction

Czech

Further Comments

Study Materials
The course is taught annually.

Bi0034 Analysis and classification of data

Extent and Intensity

2/1/0. 3 credit(s) (plus extra credits for completion). Recommended Type of Completion: zk (examination). Other types of completion: z (credit).

Teacher(s)

prof. Ing. Jiří Holčík, CSc. (lecturer)
RNDr. Eva Koriťáková, Ph.D. (lecturer)

Guaranteed by

prof. RNDr. Ladislav Dušek, Ph.D.
RECETOX – Faculty of Science
Contact Person: prof. Ing. Jiří Holčík, CSc.
Supplier department: RECETOX – Faculty of Science

Timetable

Wed 8:00–10:50 F01B1/709

Prerequisites

It is recommended to attend the course Bi8600 Multivariate Methods beforehand.

Course Enrolment Limitations

The course is offered to students of any study field.

Course objectives

At the end of the course, students should be able to: - know fundamental theoretical and methodological principles of pattern analysis and recognition with emphasis to processing of medical and biological data;- choose and apply appropriatedata analysis method to obtain required results; - reduce and transform multivariate data using advances ordination methods; - classify data using various classification methods; - design modified algorithms to process data of given particular characteristics; - interpret obtained results including evaluation of classification performance.

Syllabus

• 1. Multivariate data analysis and its aims. 2. Ordination methods – factor analysis (FA). 3. Ordination methods – canonical correspondence analysis (CCA), redundancy analysis (RDA). 4. Ordination methods – co-coordinate analysis. 5. Advanced method of feature extraction – independent component analysis (ICA) and other advanced methods. 6. Distance and similarity metrics. 7. Classification – introduction, discriminant analysis based on discriminant functions and minimal distance. 8. Classification – discriminant analysis based on boundaries: Fisher’s linear discriminant analysis. 9. Classification – discriminant analysis based on boundaries: linear support vector machines (SVM). 10. Classification – nonlinearSVM, sequential classification. 11. Evaluation of classification performance – cross-validation, comparison of classification results with random classification, comparison of classification performance of two or more classifiers. 12. Summary of methods for data analysis and classification.

Literature

• Dunham,M.H.: Data Mining: Introductory and Advanced Topics. Prentice Hall 2002
• Holčík,J.: Analýza a klasifikace signálů. [Učební texty vysokých škol] Brno, Nakladatelství VUT v Brně 1992.
• Mitchel,T.M.: Machine Learning. McGraw Hill 1997

Teaching methods

Lectures supported by Power Point presentations. Understanding of principles, methods and algorithms is emphasized. Students are continuously encouraged to interact with the lecturer.

Assessment methods

oral examination

Language of instruction

Czech

Further Comments

Study Materials
The course is taught annually.

Bi0034 Analysis and classification of data

Extent and Intensity

2/1. 3 credit(s) (plus extra credits for completion). Recommended Type of Completion: zk (examination). Other types of completion: z (credit).

Teacher(s)

prof. Ing. Jiří Holčík, CSc. (lecturer)
RNDr. Eva Koriťáková, Ph.D. (lecturer)

Guaranteed by

prof. RNDr. Ladislav Dušek, Ph.D.
RECETOX – Faculty of Science
Contact Person: prof. Ing. Jiří Holčík, CSc.
Supplier department: RECETOX – Faculty of Science

Timetable

Mon 10:00–11:50 F01B1/709, Wed 10:00–11:50 F01B1/709

Course Enrolment Limitations

The course is offered to students of any study field.

Course objectives

At the end of the course, students should be able to: - to know fundamental theoretical and methodological principles of pattern analysis and recognition with emphasis to processing of biological data - explain consequences and relationships between characteristics of real processes and data and applied methods and algorithms; - apply different practical approaches to data processing to obtain required analytic results; - design modified algorithms to process data of given particular characteristics.

Syllabus

• 1. Pattern recognition and data classification – basic vocabulary. Sorting of the classification approaches. 2. Classification based on feature selection. Classification by means of discriminant functions and minimum distance. 3. Determination of the discriminant functions based on statistical characteristics of a set of patterns. 4. Sequential classification. 5. Feature selection and extraction. 6. Principal component analysis. 7. Independent component analysis. 8. Factor analysis. 9. Determination of classificators. Methods for estimation of probability density functions and estimation of apriori probabilities of the classification categories. 10. Clustering - principles. Similarity measures. 11. Clustering methods. 12. Neural network classification.

Literature

• Holčík,J.: Analýza a klasifikace signálů. [Učební texty vysokých škol] Brno, Nakladatelství VUT v Brně 1992.
• Mitchel,T.M.: Machine Learning. McGraw Hill 1997
• Dunham,M.H.: Data Mining: Introductory and Advanced Topics. Prentice Hall 2002

Teaching methods

Lectures supported by Power Point presentations. Understanding of principles, methods and algorithms is emphasized. Students are continuously encouraged to interact with the lecturer.

Assessment methods

oral examination

Language of instruction

Czech

Further Comments

Study Materials
The course is taught annually.

Bi0034 Analysis and classification of data

Extent and Intensity

2/1. 3 credit(s) (plus extra credits for completion). Recommended Type of Completion: zk (examination). Other types of completion: z (credit).

Teacher(s)

prof. Ing. Jiří Holčík, CSc. (lecturer)
RNDr. Jiří Jarkovský, Ph.D. (lecturer)
Mgr. Terézia Černá (lecturer)
RNDr. Eva Koriťáková, Ph.D. (lecturer), RNDr. Eva Koriťáková, Ph.D. (deputy)

Guaranteed by

prof. RNDr. Ladislav Dušek, Ph.D.
RECETOX – Faculty of Science
Contact Person: prof. Ing. Jiří Holčík, CSc.
Supplier department: RECETOX – Faculty of Science

Timetable

Mon 8:00–10:50 F01B1/709

Course Enrolment Limitations

The course is offered to students of any study field.

Course objectives

At the end of the course, students should be able to: - to know fundamental theoretical and methodological principles of pattern analysis and recognition with emphasis to processing of biological data - explain consequences and relationships between characteristics of real processes and data and applied methods and algorithms; - apply different practical approaches to data processing to obtain required analytic results; - design of modified algorithms to process data of given particular characteristics.

Syllabus

• Pattern recognition and data classification – basic vocabulary. Sorting of the classification approaches. 2. Classification based on feature description. Classification by means of discriminan functions and minimum distance. 3. Determination of the discriminant function based on statistical characteristics of a set of patterns. 4. Sequential classification. 5. Feature selection and extraction. 6. Principal component analysis. 7. Independent component analysis. 8. Factor analysis. 9. Training of classificators. Methods for estimation of probability density functions and estimation of apriori probabilities of the classification categories. 10. Clustering - principles. Similarity measures. 11. Clustering methods. 12. Neural network classification.

Literature

• Dunham,M.H.: Data Mining: Introductory and Advanced Topics. Prentice Hall 2002
• Mitchel,T.M.: Machine Learning. McGraw Hill 1997
• Holčík,J.: Analýza a klasifikace signálů. [Učební texty vysokých škol] Brno, Nakladatelství VUT v Brně 1992.

Teaching methods

Lectures supported by Power Point presentations. Understanding of principles, methods and algorithms is emphasized. Students are continuously encouraged to be in an interaction with a lecturer.

Assessment methods

oral examination

Language of instruction

Czech

Further Comments

Study Materials
The course is taught annually.

Bi0034 Analysis and classification of data

Extent and Intensity

2/1. 3 credit(s) (plus extra credits for completion). Recommended Type of Completion: zk (examination). Other types of completion: z (credit).

Teacher(s)

prof. Ing. Jiří Holčík, CSc. (lecturer)
Mgr. Terézia Černá (seminar tutor)

Guaranteed by

prof. RNDr. Ladislav Dušek, Ph.D.
RECETOX – Faculty of Science
Contact Person: prof. Ing. Jiří Holčík, CSc.

Timetable

Tue 10:00–12:50 F01B1/709

Course Enrolment Limitations

The course is offered to students of any study field.

Course objectives

At the end of the course, students should be able to: - to know fundamental theoretical and methodological principles of pattern analysis and recognition with emphasis to processing of biological data - explain consequences and relationships between characteristics of real processes and data and applied methods and algorithms; - apply different practical approaches to data processing to obtain required analytic results; - design of modified algorithms to process data of given particular characteristics.

Syllabus

• Pattern recognition and data classification – basic vocabulary. Sorting of the classification approaches. 2. Classification based on feature description. Classification by means of discriminan functions and minimum distance. 3. Determination of the discriminant function based on statistical characteristics of a set of patterns. 4. Sequential classification. 5. Feature selection and extraction. 6. Principal component analysis. 7. Independent component analysis. 8. Factor analysis. 9. Training of classificators. Methods for estimation of probability density functions and estimation of apriori probabilities of the classification categories. 10. Clustering - principles. Similarity measures. 11. Clustering methods. 12. Neural network classification.

Literature

• Dunham,M.H.: Data Mining: Introductory and Advanced Topics. Prentice Hall 2002
• Mitchel,T.M.: Machine Learning. McGraw Hill 1997
• Holčík,J.: Analýza a klasifikace signálů. [Učební texty vysokých škol] Brno, Nakladatelství VUT v Brně 1992.

Teaching methods

Lectures supported by Power Point presentations. Understanding of principles, methods and algorithms is emphasized. Students are continuously encouraged to be in an interaction with a lecturer.

Assessment methods

oral examination

Language of instruction

Czech

Further Comments

Study Materials
The course is taught annually.

Bi0034 Analysis and classification of data

Extent and Intensity

2/0. 2 credit(s) (plus extra credits for completion). Recommended Type of Completion: zk (examination). Other types of completion: z (credit).

Teacher(s)

prof. Ing. Jiří Holčík, CSc. (lecturer)

Guaranteed by

prof. RNDr. Ladislav Dušek, Ph.D.
RECETOX – Faculty of Science
Contact Person: prof. Ing. Jiří Holčík, CSc.

Timetable

Wed 8:00–9:50 G2,02003

Course Enrolment Limitations

The course is also offered to the students of the fields other than those the course is directly associated with.

fields of study / plans the course is directly associated with

• Mathematical Biology (programme PřF, M-BI)

Course objectives

At the end of the course, students should be able to: - to know fundamental theoretical and methodological principles of pattern analysis and recognition with emphasis to processing of biological data - explain consequences and relationships between characteristics of real processes and data and applied methods and algorithms; - apply different practical approaches to data processing to obtain required analytic results; - design of modified algorithms to process data of given particular characteristics.

Syllabus

• Pattern recognition and data classification – basic vocabulary. Sorting of the classification approaches. 2. Classification based on feature description. Classification by means of discriminan functions and minimum distance. 3. Determination of the discriminant function based on statistical characteristics of a set of patterns. 4. Sequential classification. 5. Feature selection and extraction. 6. Principal component analysis. 7. Independent component analysis. 8. Factor analysis. 9. Training of classificators. Methods for estimation of probability density functions and estimation of apriori probabilities of the classification categories. 10. Clustering - principles. Similarity measures. 11. Clustering methods. 12. Neural network classification.

Literature

• Dunham,M.H.: Data Mining: Introductory and Advanced Topics. Prentice Hall 2002
• Mitchel,T.M.: Machine Learning. McGraw Hill 1997
• Holčík,J.: Analýza a klasifikace signálů. [Učební texty vysokých škol] Brno, Nakladatelství VUT v Brně 1992.

Teaching methods

Lectures supported by Power Point presentations. Understanding of principles, methods and algorithms is emphasized. Students are continuously encouraged to be in an interaction with a lecturer.

Assessment methods

oral examination

Language of instruction

Czech

Further Comments

Study Materials
The course is taught annually.

Bi0034 Analysis and classification of data

Extent and Intensity

2/0. 2 credit(s) (plus extra credits for completion). Recommended Type of Completion: zk (examination). Other types of completion: z (credit).

Teacher(s)

prof. Ing. Jiří Holčík, CSc. (lecturer)

Guaranteed by

prof. RNDr. Ladislav Dušek, Ph.D.
RECETOX – Faculty of Science
Contact Person: prof. Ing. Jiří Holčík, CSc.

Timetable

Wed 10:00–11:50 G2,02003

Course Enrolment Limitations

The course is also offered to the students of the fields other than those the course is directly associated with.

fields of study / plans the course is directly associated with

• Mathematical Biology (programme PřF, M-BI)

Course objectives

At the end of the course, students should be able to: - to know fundamental theoretical and methodological principles of pattern analysis and recognition with emphasis to processing of biological data - explain consequences and relationships between characteristics of real processes and data and applied methods and algorithms; - apply different practical approaches to data processing to obtain required analytic results; - design of modified algorithms to process data of given particular characteristics.

Syllabus

• Pattern recognition and data classification – basic vocabulary. Sorting of the classification approaches. 2. Classification based on feature description. Classification by means of discriminan functions and minimum distance. 3. Determination of the discriminant function based on statistical characteristics of a set of patterns. 4. Sequential classification. 5. Feature selection and extraction. 6. Principal component analysis. 7. Independent component analysis. 8. Factor analysis. 9. Training of classificators. Methods for estimation of probability density functions and estimation of apriori probabilities of the classification categories. 10. Clustering - principles. Similarity measures. 11. Clustering methods. 12. Neural network classification.

Literature

• Mitchel,T.M.: Machine Learning. McGraw Hill 1997
• Dunham,M.H.: Data Mining: Introductory and Advanced Topics. Prentice Hall 2002
• Holčík,J.: Analýza a klasifikace signálů. [Učební texty vysokých škol] Brno, Nakladatelství VUT v Brně 1992.

Teaching methods

Lectures supported by Power Point presentations. Understanding of principles, methods and algorithms is emphasized. Students are continuously encouraged to be in an interaction with a lecturer.

Assessment methods

oral examination

Language of instruction

Czech

Further Comments

Study Materials
The course is taught annually.

Bi0034 Analysis and classification of biomedical data

Extent and Intensity

2/0. 2 credit(s) (plus extra credits for completion). Recommended Type of Completion: zk (examination). Other types of completion: k (colloquium), z (credit).

Teacher(s)

prof. Ing. Jiří Holčík, CSc. (lecturer)

Guaranteed by

prof. RNDr. Ladislav Dušek, Ph.D.
RECETOX – Faculty of Science
Contact Person: prof. Ing. Jiří Holčík, CSc.

Timetable

Tue 12:00–13:50 MP2,01014a

Course Enrolment Limitations

The course is also offered to the students of the fields other than those the course is directly associated with.

fields of study / plans the course is directly associated with

• Mathematical Biology (programme PřF, M-BI)

Course objectives

The course provides students with fundamental methods and algorithms for description, analysis and classification of biomedical data. It deals with basic approaches of pattern recognition (decision-theoretic and structural approaches) and describes principles of both the approaches. Further, it discusses methods of the decision-theoretic approach in details. Classification using discrimination functions and minimum distance. Sequential classification. Choise and selection of features. Principal component analysis. Independent component analysis. Factor analysis. Classificator training. Clustering. Similarit measures. Neural networks and their application in pattern recognition. At the end of the course, students should be able to understand, design and implement the described algorithms for pattern recognition and decision-making.

Syllabus

• Pattern recognition and data classification – basic vocabulary. Sorting of the classification approaches. 2. Classification based on feature description. Classification by means of discriminan functions and minimum distance. 3. Determination of the discriminant function based on statistical characteristics of a set of patterns. 4. Sequential classification. 5. Feature selection and extraction. 6. Principal component analysis. 7. Independent component analysis. 8. Factor analysis. 9. Training of classificators. Methods for estimation of probability density functions and estimation of apriori probabilities of the classification categories. 10. Clustering - principles. Similarity measures. 11. Clustering methods. 12. Neural network classification.

Literature

• Mitchel,T.M.: Machine Learning. McGraw Hill 1997
• Dunham,M.H.: Data Mining: Introductory and Advanced Topics. Prentice Hall 2002
• Holčík,J.: Analýza a klasifikace signálů. [Učební texty vysokých škol] Brno, Nakladatelství VUT v Brně 1992.

Assessment methods

oral examination

Language of instruction

Czech

Further Comments

Study Materials
The course is taught last offered.

Bi0034 Knowledge Discovery by Machine Learning

Extent and Intensity

2/1. 3 credit(s) (plus extra credits for completion). Recommended Type of Completion: zk (examination). Other types of completion: k (colloquium), z (credit).

Teacher(s)

doc. Ing. Jan Žižka, CSc. (lecturer)
RNDr. Danka Haruštiaková, Ph.D. (assistant)

Guaranteed by

doc. Ing. Jan Žižka, CSc.
RECETOX – Faculty of Science
Contact Person: doc. Ing. Jan Žižka, CSc.

Course Enrolment Limitations

The course is also offered to the students of the fields other than those the course is directly associated with.

fields of study / plans the course is directly associated with

• Mathematical Biology (programme PřF, M-BI)

Course objectives

The subject concerns with inductive machine-learning methods using data samples. It explains algorithms, their principles, possibilities, and applications to automated non-analytic knowledge discovery in real-world data. The application capabilities are in looking for similar instances, further in classification, regression, and prediction.

Syllabus

• The relationships among data, information, and knowledge. Inductive learning. Automated knowledge discovery from information by pattern generalization. Training and testing, pattern selection and their representation. Problems connected with real data and incomplete descriptions of patterns, compensation of missing values and samples. Advanced fundamental algorithms of machine learning. Computational complexity, its approximation. Unsupervised learning (clustering) and supervised learning (classification, regression), pattern recognition. Interdisciplinary relations, application dependencies. Data preprocessing, algorithm selection, design and evaluation of experiments. Practical experiments with real data and the software system of machine-learning tools WEKA.

Literature

• Duda, R. O., Hart, P. E., Stork, D. G. (2001) Pattern Classification. Second edition. John Wiley & Sons. ISBN 0-471-05669-3

Language of instruction

Czech

Further Comments

The course is taught last offered.
The course is taught: every week.

Bi0034 Analysis and classification of data

The information about the term Autumn 2011 - acreditation is not made public

Extent and Intensity

2/1. 3 credit(s) (plus extra credits for completion). Recommended Type of Completion: zk (examination). Other types of completion: z (credit).

Teacher(s)

prof. Ing. Jiří Holčík, CSc. (lecturer)
RNDr. Jiří Jarkovský, Ph.D. (lecturer)

Guaranteed by

prof. RNDr. Ladislav Dušek, Ph.D.
RECETOX – Faculty of Science
Contact Person: prof. Ing. Jiří Holčík, CSc.

Course Enrolment Limitations

The course is offered to students of any study field.

Course objectives

At the end of the course, students should be able to: - to know fundamental theoretical and methodological principles of pattern analysis and recognition with emphasis to processing of biological data - explain consequences and relationships between characteristics of real processes and data and applied methods and algorithms; - apply different practical approaches to data processing to obtain required analytic results; - design of modified algorithms to process data of given particular characteristics.

Syllabus

• Pattern recognition and data classification – basic vocabulary. Sorting of the classification approaches. 2. Classification based on feature description. Classification by means of discriminan functions and minimum distance. 3. Determination of the discriminant function based on statistical characteristics of a set of patterns. 4. Sequential classification. 5. Feature selection and extraction. 6. Principal component analysis. 7. Independent component analysis. 8. Factor analysis. 9. Training of classificators. Methods for estimation of probability density functions and estimation of apriori probabilities of the classification categories. 10. Clustering - principles. Similarity measures. 11. Clustering methods. 12. Neural network classification.

Literature

• Dunham,M.H.: Data Mining: Introductory and Advanced Topics. Prentice Hall 2002
• Mitchel,T.M.: Machine Learning. McGraw Hill 1997
• Holčík,J.: Analýza a klasifikace signálů. [Učební texty vysokých škol] Brno, Nakladatelství VUT v Brně 1992.

Teaching methods

Lectures supported by Power Point presentations. Understanding of principles, methods and algorithms is emphasized. Students are continuously encouraged to be in an interaction with a lecturer.

Assessment methods

oral examination

Language of instruction

Czech

Further Comments

The course is taught annually.
The course is taught: every week.

Bi0034 Analysis and classification of data

Extent and Intensity

2/0. 2 credit(s) (plus extra credits for completion). Recommended Type of Completion: zk (examination). Other types of completion: z (credit).

Teacher(s)

prof. Ing. Jiří Holčík, CSc. (lecturer)

Guaranteed by

prof. RNDr. Ladislav Dušek, Ph.D.
RECETOX – Faculty of Science
Contact Person: prof. Ing. Jiří Holčík, CSc.

Course Enrolment Limitations

The course is also offered to the students of the fields other than those the course is directly associated with.

fields of study / plans the course is directly associated with

• Mathematical Biology (programme PřF, M-BI)

Course objectives

At the end of the course, students should be able to: - to know fundamental theoretical and methodological principles of pattern analysis and recognition with emphasis to processing of biological data - explain consequences and relationships between characteristics of real processes and data and applied methods and algorithms; - apply different practical approaches to data processing to obtain required analytic results; - design of modified algorithms to process data of given particular characteristics.

Syllabus

• Pattern recognition and data classification – basic vocabulary. Sorting of the classification approaches. 2. Classification based on feature description. Classification by means of discriminan functions and minimum distance. 3. Determination of the discriminant function based on statistical characteristics of a set of patterns. 4. Sequential classification. 5. Feature selection and extraction. 6. Principal component analysis. 7. Independent component analysis. 8. Factor analysis. 9. Training of classificators. Methods for estimation of probability density functions and estimation of apriori probabilities of the classification categories. 10. Clustering - principles. Similarity measures. 11. Clustering methods. 12. Neural network classification.

Literature

• Dunham,M.H.: Data Mining: Introductory and Advanced Topics. Prentice Hall 2002
• Mitchel,T.M.: Machine Learning. McGraw Hill 1997
• Holčík,J.: Analýza a klasifikace signálů. [Učební texty vysokých škol] Brno, Nakladatelství VUT v Brně 1992.

Teaching methods

Lectures supported by Power Point presentations. Understanding of principles, methods and algorithms is emphasized. Students are continuously encouraged to be in an interaction with a lecturer.

Assessment methods

oral examination

Language of instruction

Czech

Further Comments

The course is taught annually.
The course is taught: every week.

Bi0034 Knowledge Discovery by Machine Learning

Extent and Intensity

2/1. 3 credit(s) (plus extra credits for completion). Recommended Type of Completion: zk (examination). Other types of completion: k (colloquium), z (credit).

Teacher(s)

doc. Ing. Jan Žižka, CSc. (lecturer)
RNDr. Danka Haruštiaková, Ph.D. (assistant)

Guaranteed by

doc. Ing. Jan Žižka, CSc.
RECETOX – Faculty of Science
Contact Person: doc. Ing. Jan Žižka, CSc.

Course Enrolment Limitations

The course is also offered to the students of the fields other than those the course is directly associated with.

fields of study / plans the course is directly associated with

• Mathematical Biology (programme PřF, M-BI)

Course objectives

The subject concerns with inductive machine-learning methods using data samples. It explains algorithms, their principles, possibilities, and applications to automated non-analytic knowledge discovery in real-world data. The application capabilities are in looking for similar instances, further in classification, regression, and prediction.

Syllabus

• The relationships among data, information, and knowledge. Inductive learning. Automated knowledge discovery from information by pattern generalization. Training and testing, pattern selection and their representation. Problems connected with real data and incomplete descriptions of patterns, compensation of missing values and samples. Advanced fundamental algorithms of machine learning. Computational complexity, its approximation. Unsupervised learning (clustering) and supervised learning (classification, regression), pattern recognition. Interdisciplinary relations, application dependencies. Data preprocessing, algorithm selection, design and evaluation of experiments. Practical experiments with real data and the software system of machine-learning tools WEKA.

Literature

• Duda, R. O., Hart, P. E., Stork, D. G. (2001) Pattern Classification. Second edition. John Wiley & Sons. ISBN 0-471-05669-3

Language of instruction

Czech

Further Comments

The course is taught annually.
The course is taught: every week.

• Enrolment Statistics (recent)