University of Nevada - Reno
Exploration of the Electroencephalogram Signal Data with Functional Principal Components
Abstract
dc:description.abstractThis thesis investigates the analysis and classification of electroencephalogram (EEG) signals using both multivariate and functional data analysis techniques. The EEG signals are recorded using two different electrode technologies: emulated and tripolar EEG. The subjects of the study were given a visual task involving checkerboard stimuli and varying gaze directions. The data consist of high-frequency time series recordings, naturally treated within a functional data framework. The primary objective was to determine whether gaze direction, left versus right fixation, can be reliably classified basedon the EEG signals. The data consist of high-frequency time series recordings, naturally treated within a functional data framework. To address the high dimensionality of the data, both Multivariate Principal Component Analysis and Functional Principal Component Analysis were employed for dimension reduction, followed by classification using methods such as k-means clustering, naive classifiers, and Support Vector Machines (SVMs). The results indicate that both signal types pose challenges for reliably distinguishing subtle differences, such as gaze direction. Classification performance varies across methods and data representations, with SVMs generally outperforming other approaches. Overall, the findings suggest that although advanced electrode designs and functional methods enhance analysis, the discrimination of closely related neural signals remains challenging.
Degree
thesis:*- Level thesis:degree_level
- Master’s Degree
- Department thesis:degree_department
- Mathematics
- Year dc:date.issued
- 2026
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Labenski, Tanja
- Advisor dc:contributor.advisor
-
- Panorska, Anna K
- Committee members dc:contributor.committeemember
-
- Kozubowski, Tomasz J
- Crognale, Michael A
- Kemmelmeier, Markus
Subjects
dc:subject × 3Rights
- Language dc:language.iso
- en_US, English
Identifiers
dc:identifier.*- Repository record dc:identifier.uri
- https://scholarwolf.unr.edu/handle/11714/11919