FRITZ, K. Emotion Recognition from Brain Electroencephalogram (EEG) Signals [online]. Brno: Vysoké učení technické v Brně. Fakulta informačních technologií. 2023.

Posudky

Posudek vedoucího

Malik, Aamir Saeed

Overall, the work was satisfactory. 

Dílčí hodnocení
Kritérium Známka Body Slovní hodnocení
Informace k zadání The thesis was related to emotion detection from the EEG data. In general, EEG data analysis is challenging because of the large number of electrodes, correlation amongthe electrodes and the variation in data with time. The work done is satisfactory keeping in mind the complexity of the topic.
Práce s literaturou The student worked hard in the first semester to do the literature review and to find the gap in the existing methods. He struggled to come up with the proposed method to provide the solution. Overall, he worked hard and tried his best.
Aktivita během řešení, konzultace, komunikace The student had regular meetings with me, generally once a wekk or every two weeks. Most of the times, he was prepared for the consultation. Overall, he tried to meet the deadlines set in the meetings.
Aktivita při dokončování The student had to extend the time duration from May to July for the submission of the thesis. This was due to the registration of quite a number of courses in the summer semester. I personally feel that students should not register for more than 5 courses in a semster.
Publikační činnost, ocenění It is possible to write a conference paper out of the work done by the student. I will encourage the student to prepare the draft for the conference paper.
Navrhovaná známka
D
Body
69

Posudek oponenta

Jawed, Soyiba

The idea of this work is innovative and has a lot of potential for further improvement. It is a satisfactory work based on state-of-the-art techniques for neuroimaging analysis using deep learning, which leads to the originality of the thesis. The work demonstrates the author's deep interest in the neural network and the ability to analytical thinking. Except for minor shortcomings in the representation of results, the work is well executed in terms of structure and language. Further improvement can be made for the reusability in future.

Dílčí hodnocení
Kritérium Známka Body Slovní hodnocení
Náročnost zadání A demanding research assignment required an understanding and experimental mastery of signal processing and neural networks applied to EEG dataset.
Rozsah splnění požadavků zadání The assignment was fulfilled, but some modifications are required in the thesis in terms of graphs and references. He did time and feature extraction and classification of emotions using an EEG dataset.
Rozsah technické zprávy The work is within the usual range. It explains and covers almost all the necessary points. But it would be better if some additional information is added such as more explanation of the mathematics behind the neural networks.
Prezentační úroveň technické zprávy 70 The thesis is written following the BUT format. The presentation is standard. Besides some text is not referenced. The flow is there but improvement can be made. 
Formální úprava technické zprávy 75 Formally, the work is very nicely executed in a readable manner. The language level is satisfactory, and the work is written in good English with minimal errors. The figures and table are well arranged.
Práce s literaturou 75 The work with literature follows the standard. The student read and gave an overview of the state of the art. Besides that, some of the work requires citations.  
Realizační výstup 76 The output of the work is a set of scripts using Python language libraries for deep learning (pytorch) and EEG modality. The results are presented in readable manner.
Využitelnost výsledků The thesis has a strong potential to use further in the area of emotion detection. With some tweaking, it could also be published in some reputable journal.
Navrhovaná známka
C
Body
75

eVSKP id 141164