BLÁHA, J. EEG-based classification of anxiety subtypes using machine learning [online]. Brno: Vysoké učení technické v Brně. Fakulta informačních technologií. 2025.
This thesis constitutes a well-executed and practical bachelor-level study focused on the application of EEG data and machine learning techniques to classify different types of anxiety. The student demonstrated strong theoretical knowledge and technical skills, resulting in a functional prototype capable of processing and analyzing relevant data. Overall, the thesis is a commendable and effective contribution that meets the academic standards expected at the bachelor’s level.
| Kritérium | Známka | Body | Slovní hodnocení |
|---|---|---|---|
| Informace k zadání | This thesis focused on the development of a digital tool designed to support the classification of anxiety using basic machine learning techniques. The aim was to explore how digital environments can be utilized to identify patterns potentially associated with anxiety, using computational models. The project involved designing the assessment framework, applying introductory-level classification methods to evaluate its potential in distinguishing individuals based on anxiety-related traits. The work required a foundational understanding of anxiety from a psychological perspective, alongside technical skills in programming, data handling, and machine learning. While the approach was exploratory, it demonstrated how digital tools can contribute to early identification or monitoring of anxiety through automated analysis. Overall, the thesis presented a balanced integration of psychological insight and computational methods. It was suitably challenging for undergraduate research and resulted in a working prototype that reflects both theoretical understanding and practical implementation of anxiety classification using digital data. | ||
| Práce s literaturou | As the supervisor, I observed that the student was consistently engaged and dedicated throughout the thesis process. We held regular meetings—typically once a week—to discuss progress, clarify concepts, and review relevant literature. These sessions provided a platform for in-depth discussions and effective guidance. In addition to our meetings, the student frequently reached out via email whenever further input or clarification was needed, demonstrating strong initiative and a proactive approach to learning. I recommended key academic resources to support their understanding of anxiety assessment using EEG data and machine learning methods, which the student actively explored and applied. Overall, the student showed commendable independence and a strong commitment to producing quality work. | ||
| Aktivita během řešení, konzultace, komunikace | The student demonstrated consistent dedication and active engagement throughout the thesis process. He reliably met deadlines and maintained clear, timely communication about his progress, reflecting a strong sense of responsibility expected at the higher education level. His communication was professional and receptive to feedback, showing a genuine eagerness to learn and improve. This proactive attitude, combined with his efforts to effectively apply EEG data analysis and machine learning techniques to anxiety research, significantly contributed to the successful and timely completion of his thesis. | ||
| Aktivita při dokončování | The final thesis underwent thorough review and refinement through several rounds of feedback. He welcomed constructive criticism and diligently incorporated revisions that enhanced the clarity, flow, and depth of the work—particularly in sections addressing EEG analysis, anxiety assessment, and the application of machine learning techniques. | ||
| Publikační činnost, ocenění | The student has not yet initiated any formal publication efforts related to this bachelor’s thesis. However, the results obtained are promising and show strong potential. Further refinement and improvement of the preprocessing will be necessary before it is ready for publication. |
Overall, the student has done good work in the thesis. The chapter on literature review is very good and the student has provided overview of the published articles. The proposed methodology is based on features extraction and correspondingly building an Ensemble Learning Model. The implementation and the results are provided in detail. However, the main shortcoming is in the pre-processing step which could have been improved.
| Kritérium | Známka | Body | Slovní hodnocení |
|---|---|---|---|
| Náročnost zadání | It is a difficult project because it requires not only knowledge in computer science but also needs to incorporate psychological, psychiatric and neurological concepts of anxiety. The challenges in understanding the underlying brain mechanisms, and the corresponding features extraction as well as their interpretation with respect to anxiety, makes it a difficult project. | ||
| Rozsah splnění požadavků zadání | The thesis is well structured and well written. Overall, it is a good thesis because the student did study the problem, conducted a detailed literature review, proposed a model for the problem, implemented it, and provided the results with good number of quality metrics. The main shortcoming of the thesis is in the pre-processing step which also affect the results of the thesis. | ||
| Rozsah technické zprávy | The thesis has about 70 pages and hence meets the minimum 40 standard pages requirement. The student has done a good literature review and then provided sufficient details for the proposed methodology, implementation and results. | ||
| Prezentační úroveň technické zprávy | 85 | Overall, the structure of the thesis, in terms of chapters organization, is good. Chapter 1 provides an introduction, followed by theoretical introduction and literature review in chapters 2 and 3. Chapters 4 and 5 provide details of the proposed method and the corresponding implementation of the proposed method. Results are discussed in chapter 6 followed by conclusion in chapter 7. A summary section is provided at the end of each chapter. Student has effectively used sections and sub-sections for providing details in the relevant chapters. | |
| Formální úprava technické zprávy | 80 | In terms of language, the thesis is readable and use of English language in the thesis is good. There are minor typos and grammatical mistakes. However, they are few and in general the thesis is well written. | |
| Práce s literaturou | 84 | The student has done a good literature review. It is a very well written chapter with a detailed description of the published work. In addition, the student has provided nice tables at the end of every section highlighting the summary of the important articles with their limitations. | |
| Realizační výstup | 72 | Proposed methodology is well written with details of all the steps provided in chapter 4. The details of the datasets are provided in section 4.1. Sections 4.2 to 4.9 provides details of each step of the proposed methodology. However, student preprocessing part can be significantly improved. The implementation is provided in detail in chapter 5. The student has addressed topics like labelling and oversampling. The student uses Ensemble Learning which is quite a good approach for this problem with multiclass classification. Again, the results are provided in detail in chapter 6. Comparison of the results with other methods is not provided. The main shortcoming of the methodology is in the pre-processing step which also affect the results of the thesis. | |
| Využitelnost výsledků | The work consists of utilizing the traditional machine learning method of feature extraction and selection. The student proposes to use Ensemble Learning approach. The novelty is the combination of all these features and then providing the significant ones to the classifier. The student provides sufficient details in all the chapters. |
eVSKP id 161455