Statistical and machine learning methods for classification of episodic memory
(2018)Department of Automatic Control
- Abstract
- Multiple modern methods of statistical feature extraction and machine learning are applied to classification of encoding and retrieval of episodic memories using electroencephalogram (EEG) recordings. Raw data, different time-frequency methods, and multiclass common spatial patterns are used for statistical feature extraction. For each type of feature extraction multiple machine learning algorithms are tested and compared. Classification accuracies of up to 82 % are reached with one-dimensional convolutional neural networks on raw data. It is found that more complex and time-consuming classifiers generally improve the accuracy. However, the features chosen are the main factor deciding the accuracy. A novel idea for designing an... (More)
- Multiple modern methods of statistical feature extraction and machine learning are applied to classification of encoding and retrieval of episodic memories using electroencephalogram (EEG) recordings. Raw data, different time-frequency methods, and multiclass common spatial patterns are used for statistical feature extraction. For each type of feature extraction multiple machine learning algorithms are tested and compared. Classification accuracies of up to 82 % are reached with one-dimensional convolutional neural networks on raw data. It is found that more complex and time-consuming classifiers generally improve the accuracy. However, the features chosen are the main factor deciding the accuracy. A novel idea for designing an encoding-retrieval classifier is discussed and implemented. In spite of having multiple different designs, almost all classifier combinations involving the retrieval data fail to reach significant classification levels. (Less)
Please use this url to cite or link to this publication:
http://lup.lub.lu.se/student-papers/record/8953710
- author
- Basic Knezevic, Damir and Heimerson, Albin
- supervisor
-
- Bo Bernhardsson LU
- Anton Cervin LU
- organization
- year
- 2018
- type
- H3 - Professional qualifications (4 Years - )
- subject
- report number
- TFRT-6056
- ISSN
- 0280-5316
- language
- English
- id
- 8953710
- date added to LUP
- 2018-06-29 12:16:50
- date last changed
- 2018-06-29 12:16:50
@misc{8953710,
abstract = {{Multiple modern methods of statistical feature extraction and machine learning are applied to classification of encoding and retrieval of episodic memories using electroencephalogram (EEG) recordings. Raw data, different time-frequency methods, and multiclass common spatial patterns are used for statistical feature extraction. For each type of feature extraction multiple machine learning algorithms are tested and compared. Classification accuracies of up to 82 % are reached with one-dimensional convolutional neural networks on raw data. It is found that more complex and time-consuming classifiers generally improve the accuracy. However, the features chosen are the main factor deciding the accuracy. A novel idea for designing an encoding-retrieval classifier is discussed and implemented. In spite of having multiple different designs, almost all classifier combinations involving the retrieval data fail to reach significant classification levels.}},
author = {{Basic Knezevic, Damir and Heimerson, Albin}},
issn = {{0280-5316}},
language = {{eng}},
note = {{Student Paper}},
title = {{Statistical and machine learning methods for classification of episodic memory}},
year = {{2018}},
}