460-4139/01 – Machine Learning (SU)
| Gurantor department | Department of Computer Science | Credits | 4 |
| Subject guarantor | prof. Ing. Jan Platoš, Ph.D. | Subject version guarantor | prof. Ing. Jan Platoš, Ph.D. |
| Study level | undergraduate or graduate | Requirement | Optional |
| Year | 2 | Semester | winter |
| | Study language | Czech |
| Year of introduction | 2022/2023 | Year of cancellation | |
| Intended for the faculties | FEI | Intended for study types | Follow-up Master |
Subject aims expressed by acquired skills and competences
The course aims to provide students with a detailed overview of procedures and methods in machine learning, from exploratory data analysis, through the search for similarity, comparison of objects to the search for classification models. Students will have the chance to implement and test individual methods on artificial and real data and evaluate the results they will learn to present correctly.
Teaching methods
Lectures
Tutorials
Teaching by an expert (lecture or tutorial)
Summary
In the course, students get acquainted with the properties of data, their storage, and processing. They will also get acquainted with data analysis methods, machine learning, artificial intelligence, interpretation of results, and visualization. Lectures will focus on basic methods of analysis and data and extraction of findings extracted from data. Students will decide for themselves when which method is suitable, its assumptions, what its principle is, and what outputs can be obtained with it. The exercise will then be used for practical experiments on suitable data sets, experimentation with tools for data analysis, and evaluation of results.
Compulsory literature:
Recommended literature:
Additional study materials
Way of continuous check of knowledge in the course of semester
The student knowledge is checked during lab using exercises, home works and implementation of selected algorithm.
E-learning
All materials are published on the e-learning portal (https://www.vsb.cz/e-vyuka/en).
Other requirements
Additional requirements are not placed on the student.
Prerequisities
Subject has no prerequisities.
Co-requisities
Subject has no co-requisities.
Subject syllabus:
Lectures (topics):
1. Frequent patterns in data.
2. Exploratory data analysis.
3. Representative clustering, Hierarchical clustering.
4. Clustering based on data density, cluster validation.
5. Special clustering methods, detection of outliers.
6. Linear classifiers (Linear discriminant analysis, Naive Bayes, Logistic regression).
7. Decision trees, rule classification.
8. Support Vector Machine, Kernel methods.
9. Neural networks.
10. Regression methods and Advanced methods in data classification.
11. Validation of classification algorithms.
12. Time series analysis.
Exercises in the computer room (topics):
1. Implementation of the APRIORI method for searching for rules in data.
2. Exploratory analysis of data over a real dataset using appropriate tools.
3. Implementation of hierarchical clustering - Agglomerative clustering.
4. Implementation of DBSCAN algorithm.
5. Real example of clustering - independent work on exercises.
6. Dimension reduction.
7. Implementation of decision tree.
8. Testing the Support Vector Machine method over real data.
9. Neural networks.
10. Regression methods.
11. Ensemble methods and their use.
12. Classification - real example.
13. Time series analysis.
Conditions for subject completion
Occurrence in study plans
Occurrence in special blocks
Assessment of instruction