Course detail

Multi-valued Logic Applications

FSI-SALAcad. year: 2023/2024

The course is designed for students of mathematical engineering and contains the theory of fuzzy logic, linguistic variables and linguistic models and the theory of expert systems. The subject also includes the practical design of an expert system based on Lukasiewicz or Mamdani logic.

The second part of the course is devoted to machine learning and neural networks, which are used for modern applications of expert systems. Students become familiar with basic terminology, other types and their use for applications (speech, image, etc. analysis).

Language of instruction

Czech

Number of ECTS credits

4

Mode of study

Not applicable.

Guarantor

prof. RNDr. Miloslav Druckmüller, CSc.

Department

Institute of Mathematics (ÚM)

Entry knowledge

Mathematical logic, Fuzzy set theory.

Rules for evaluation and completion of the course

Graded assessment based on submission of semester work (70 percent) and oral exam of the given theory (30 percent).


Participation on lessons is compulsory, in case of absence it is necessary to work out substitute work.

Aims

The aim of the course is to introduce the methods of fuzzy logic and the proposition of expert systems. Next, students will learn to design a simple system based on machine learning and will learn the theoretical and practical basics of neural networks.


1.Terminology and explanation of the concepts of many-valued logic.
2. ntroducing word models, designing an expert system.
3. Machine learning methods.
4. Neural networks (NN) - basic properties and concepts.
5. Use of NN for analysis of text, speech, image (CNN). Design your own neural network without even using pre-trained models.

Study aids

Not applicable.

Prerequisites and corequisites

Not applicable.

Basic literature

Druckmüller, M.: Technické aplikace vícehodnotové logiky, PC- DIR , Brno 1998 (CS)

Recommended reading

Neural Networks and Deep Learning. Online. Michael Nielsen, 2015. Dostupné z: http://neuralnetworksanddeeplearning.com/. [cit. 2023-10-30]. (EN)

Classification of course in study plans

  • Programme N-MAI-P Master's 1 year of study, summer semester, elective

Type of course unit

 

Lecture

26 hod., compulsory

Teacher / Lecturer

prof. RNDr. Miloslav Druckmüller, CSc.

Syllabus

1. Multi-valued logic, formulae.
2. T-norms, T-conorms, generalized implications.
3. Linguistic variables and linguistic models, knowledge bases of expert systems.
4. Lukasiewicz logic, Mamdani principle.
5. Inference techniques and its implementation, redundance and contradictions in knowledge bases, Fuzzification and defuzzification problem.
6. Expert system project.
7.Machine learning (decisiton trees, kNN, SVM).
8.-9. Text mining, chatbot
10. Neural network elementary principles, Deep Learning.
11. Convolutional Neural Network (CNN).
12.-13. Semestral work, consultation. 

Computer-assisted exercise

13 hod., compulsory

Teacher / Lecturer

prof. RNDr. Miloslav Druckmüller, CSc.

Syllabus

Topics for work in exercises are closely related to the lectures. As part of the computer exercises, particular areas will be implemented in Matlab software, event. Python. IBM Watson Assistant will be used to design the chatbot. 

1. Multi-valued logic, formulae.
2. T-norms, T-conorms, generalized implications.
3. Linguistic variables and linguistic models, knowledge bases of expert systems.
4. Lukasiewicz logic, Mamdani principle.
5. Inference techniques and its implementation, redundance and contradictions in knowledge bases, Fuzzification and defuzzification problem.
6. Expert system project.
7.Machine learning (decisiton trees, kNN, SVM).
8.-9. Text mining, chatbot
10. Neural network elementary principles, Deep Learning.
11. Convolutional Neural Network (CNN).
12.-13. Semestral work, consultation.