MIKOGAZIEV, M. Utilizing electroencephalogram (EEG) data alongside machine learning for the classification of different levels of mental stress [online]. Brno: Vysoké učení technické v Brně. Fakulta informačních technologií. 2025.
This thesis presents a well-executed and practical bachelor-level project focused on classifying different levels of stress using EEG data and basic machine learning techniques. The student demonstrated a solid grasp of both the theoretical and technical aspects, resulting in a working prototype capable of handling and analyzing relevant data. While the work is commendable and meets the academic expectations for a bachelor’s thesis, there is still considerable room for improvement—particularly in refining the written content, expanding the theoretical discussion, and enhancing the overall presentation. Even so, it serves as a strong foundation that can be further developed in future research.
| Kritérium | Známka | Body | Slovní hodnocení |
|---|---|---|---|
| Informace k zadání | This thesis focused on developing a digital tool aimed at identifying and classifying four distinct levels of stress. The goal was to explore how computational methods can help recognize patterns associated with varying intensities of stress through digital data. By designing an assessment framework and applying basic machine learning techniques, the project sought to evaluate the tool’s ability to distinguish between different stress levels effectively. This approach offers a novel way to understand stress by combining psychological concepts with computational analysis. The work required foundational knowledge of stress from a psychological standpoint, along with essential skills in programming and data handling. The results demonstrated that digital tools, when paired with machine learning, can support the early detection and monitoring of stress in an automated and scalable way. Overall, the thesis presented a balanced integration of theory and practical implementation. It was well-suited for undergraduate research and resulted in a functional prototype that reflects both the conceptual understanding of stress and its application through digital technology. | ||
| Práce s literaturou | As a supervisor, I noticed that the student stayed really engaged and committed throughout the whole thesis process. We had regular meetings, usually once a week, where we talked about progress, cleared up any doubts, and went through important readings together. These meetings were helpful and allowed us to have good, meaningful discussions. Besides that, the student often reached out by email whenever they needed extra help or had questions, which showed they were eager to learn and took initiative. I suggested some key resources on stress assessment with EEG and machine learning, and the student made a solid effort to understand and use them. That said, while the student worked hard and showed independence, there’s still room to improve, especially in digging deeper into the material and polishing some technical skills. With a bit more practice and focus, I’m confident they can get even better. Overall, the student did a good job and built a strong foundation to build on. | ||
| Aktivita během řešení, konzultace, komunikace | The student showed consistent dedication and stayed actively engaged throughout the thesis work. He was well-prepared for meetings, often coming in with his own research and questions, which led to meaningful and productive discussions. This demonstrated not only responsibility but also a genuine interest in the topic. He reliably met deadlines and communicated his progress clearly and on time. His professional attitude and openness to feedback reflected a strong willingness to learn and improve. The student’s proactive approach, combined with his solid efforts to apply EEG data analysis and machine learning techniques, played a key role in the successful and timely completion of his thesis. | ||
| Aktivita při dokončování | The final thesis was reviewed and revised, though the refinement process was somewhat limited as the student continued working on the content until the later stages of the timeline. As a result, we were able to go through only two rounds of feedback. Despite the time constraints, he responded well to the comments provided and made thoughtful revisions that improved the structure and clarity—especially in the sections related to EEG analysis, stress assessment, and the use of machine learning techniques. | ||
| Publikační činnost, ocenění | The student has not yet taken any formal steps toward publishing the bachelor’s thesis. While the results are promising and show clear potential, the thesis still requires considerable work in terms of refining the written content. The main focus should be on improving the clarity and depth of the theoretical sections, as the core results themselves are relatively sound and may not need major changes. With thoughtful revision, especially on the conceptual side, the work could move closer to being publication-ready. |
Overall, the work presented in the thesis meets the requirements for a Bachelor thesis. The student studied the problem at hand, proposed machine learning model for solving the problem, implemented them, and provided the results with enough quality metrics. Comparison and visualization of the results could have been better. The main shortcoming of the methodology is in the pre-processing step which also affect the results of the thesis.
| 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 mental stress. The challenges in understanding the underlying brain mechanisms, and the corresponding features extraction as well as their interpretation with respect to mental stress, makes it a difficult project. | ||
| Rozsah splnění požadavků zadání | Overall, it is satisfactory thesis because the student studied the problem, conducted a literature review, proposed a model for the problem, implemented it, and provided the results with good number of quality metrics. The shortcoming is in the literature review where the student did not build hypothesis based on gaps that he identified in the review. The main shortcoming of the methodology is in the pre-processing step which also affect the results of the thesis. | ||
| Rozsah technické zprávy | The thesis meets the minimum 40 standard pages requirement. Though it meets the requirement, the student could have easily expanded to more than 40 pages by providing details of limitations and gaps and correspondingly building a hypothesis. | ||
| Prezentační úroveň technické zprávy | 78 | Overall, the structure of the thesis, in terms of chapters organization, is satisfactory. Chapter 1 provides an introduction, followed by brain anatomy and physiology in chapter 2. Chapters 3 provides literature review while 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. Some of the chapters end suddenly. There should be a summary section at the end of the chapters. Also, some figures and tables are not referenced in the text. | |
| Formální úprava technické zprávy | 80 | In terms of language, the thesis is readable and use of English language in the thesis is satisfactory. There are minor typos and grammatical mistakes. However, they are few and in general the thesis is fine. | |
| Práce s literaturou | 77 | Overall, the literature review is satisfactory. In chapter 2, the details of neuroanatomy and neurophysiology are provided while chapter 3 provides detailed literature review of EEG studies with respect to mental stress. The shortcoming in the literature review is where the student did not build hypothesis based on gaps identified in review. | |
| Realizační výstup | 72 | The proposed methodology is satisfactory. The main shortcomings are the justifications for the various features selected as well as some of the pre-processing steps. The results are provided in detail. However, the clarity is not there in the results in terms of features selection. It would have been good to provide topomaps and other feature visualization graphs. Finally, comparison with other methods is not provided. | |
| Využitelnost výsledků | The thesis provides method for the classification of mental stress, that is, traditional machine learning approach using ensemble learning with feature extraction. The student did not provide comparison with state-of-the-art in assessment of mental stress. Hence, it is not possible to assess the possibility of using the results in practice, though it is possible in general for such a thesis. |
eVSKP id 161428