Knowledge acquisition by computational intelligence
- Credits: 4
- Ending: Examination
- Range: 2P + 2C
- Semester: summer
- Faculty of Economic Informatics
Teachers
Included in study programs
Teaching results
Learning Outcomes after Course Completion
Knowledge
A. Understand the importance of data preparation for knowledge discovery tasks.
B. Understand the significance of the four main data mining problems: classification, clustering, association rule mining, and detection of extreme or outlier values.
C. Understand the principles of computational intelligence and machine learning and their application contexts.
D. Know software environments and programming languages used in computational intelligence and machine learning.
F. Gain an overview of the significance, benefits, and challenges of computational intelligence and machine learning.
G. Understand approaches to achieving explainability of results produced by data mining and machine learning models.
Skills
E. Logically aggregate atomic requirements for the purpose of knowledge discovery from data.
H. Apply acquired knowledge and skills to solving practical data mining and knowledge discovery tasks.
I. Work independently and collaboratively with selected software tools for knowledge discovery from data.
Competences
H. Independently and responsibly apply data mining, computational intelligence, and machine learning methods when solving real-world problems.
I. Effectively collaborate in a team when solving complex knowledge discovery tasks and take responsibility for the quality of analytical outcomes.
Indicative content
1. Introduction into data mining (principles, explaining the basics and relevance of data mining in knowledge discovery form data).
2. Introduction into computational intelligence and machine learning for the purpose of interpreting knowledge form data.
3. Data types, their properties and conversion.
4. Distance and similarity metrics, normalization methods in data preparation in extracting knowledge from data.
5. The key areas in data mining: clustering, associative rules, outlier detection and classification.
6. Exploring the MATLAB and WEKA environment, working with the language Python.
7. Python in tasks from data preparation to visualization of results.
8. Types of neural networks and classification by neural networks.
9. Modelling rule-based systems and inference in discovering knowledge from data mining.
10. Evolutionary algorithms and their applicability.
11. Modeling recommender systems based on customers requirements, similarities among items and historical data
12. Interactive acquisition of knowledge from data (environments for unsupervised data mining)
13. Automatic acquisition of knowledge from data (environments for supervised data mining)
Support literature
BRAMER, Max. Principles of data mining. London: Springer-Verlag, 2020.
AGGRAWAL, C. Data Mining: The Textbook. Cham: Springer, 2015
BERKA, Petr. Dobývání znalostí z databází. Praha: Academia, 2003.
SKALSKÁ, Hana. Data mining a klasifikační modely. Hradec Králové: Gaudeamus, 2010.
BRUNTON, S.L., KUTZ, J.N., Data-Driven Science and Engineering. Machine learning, Dynamical Systems and Control, Cambridge University press, 2019
NEGNEVITSKY, M. Artificial Intelligence A Guide to Intelligent Systems, Pearson, 2011
MARČEK, D., Neurónové siete a ich aplikácie, EDIS, 2006
NÁVRAT, P. a kol., Umelá inteligencia, STU, 2011
CHOLLET, F., Deep learning v jazyku Python, GRADA, 2019
PECINOVSKÝ, R., Python - Kompletní příručka jazyka pro verzi 3.8, GRADA, 2019
Syllabus
1. The introductory part focuses on data mining, its fundamental principles, terminology, and objectives. The importance of data mining in the process of knowledge discovery from data is emphasized. 2. The course introduces computational intelligence and machine learning in the context of acquiring and interpreting knowledge from data. Their relationship to data mining and decision support is explained. 3. Attention is given to data types and their properties. Data transformation and conversion techniques are discussed as part of data preparation for knowledge discovery tasks. 4. Distance and similarity measures used in data mining are introduced. The importance of data normalization and the methods used in data preprocessing are emphasized. 5. Key areas of data mining are presented, including clustering, association rule mining, classification, and detection of extreme and isolated values. Their purpose and application contexts are discussed. 6. The course explores software environments MATLAB and WEKA. Demonstrated is solving selected data mining tasks using these tools. 7. The syllabus covers working with the Python programming language in data mining. The complete process from data preparation to evaluation of data mining results is discussed. 8. Types of neural networks and their application in classification tasks are presented. Principles of neural network learning and the interpretation of results are emphasized. 9. Rule-based system modeling and inference using rule-based systems are discussed. Their significance in extracting knowledge from data is explained. 10. Evolutionary algorithms and their use in machine learning are introduced. Fundamental principles of evolutionary computation and application areas are discussed. 11. The course addresses modeling of recommender systems based on customer requirements, product or service similarity, and historical data. Their role in decision support is emphasized. 12. Automatic knowledge discovery from data is discussed with emphasis on creating environments for data mining without direct user intervention. 13. The final part focuses on interactive knowledge discovery from data. The creation of environments in which users actively participate in and influence the data mining process is discussed.
Requirements to complete the course
Exam 60% The exam consists of two parts: the evaluation of the theoretical knowledge and knowledge of modelling particular tasks. The first part, verifies the achievement level of the teaching results A., B., C., F., whereas the second part by the solving tasks verifies the level of the teaching results E., G.
Assignments during the semester 40% The purpose of seminars is designing and defending the project. Students are cooperating in small groups on projects. The evaluation of the students involves achievement in the project, answers to the supplementary questions and short test. The evaluation assesses the following teaching results: A., D., E., G., H., I.
Student workload
Total study load (in hours):
4 credits x 52 teaching hours = 130 h
Distribution of study load:
lectures and seminars participation: 52 h
seminar participation: 13 h
project and test preparation: 30 h
preparation of exam: 35 h
Language whose command is required to complete the course
anglický
Date of approval: 04.03.2025
Date of the latest change: 04.01.2026