TY - GEN
T1 - Dendrite Ellipsoidal Neuron Trained by Stochastic Gradient Descent for Motor Imagery Classification
AU - Arce, Fernando
AU - Mendoza-Montoya, Omar
AU - Zamora, Erik
AU - Antelis, Javier M.
AU - Sossa, Humberto
AU - Cantillo-Negrete, Jessica
AU - Carino-Escobar, Ruben I.
AU - Hernández, Luis G.
AU - Falcón, Luis Eduardo
N1 - Publisher Copyright:
© 2019, Springer Nature Switzerland AG.
PY - 2019
Y1 - 2019
N2 - Dendrite ellipsoidal neurons are a novel and different alternative for classification tasks, giving competitive results compared with typical classification methods. Based on k-means++ algorithm, the network allows each dendrite to build a hyperellipsoidal in order to assign each incoming pattern to its respective C class. The main disadvantage of this training algorithm is the lack of accuracy in high dimensional datasets. In this research, we solved this problem by training the dendrite ellipsoidal neuron using stochastic gradient descent. Furthermore, electroencephalography data were acquired during two mental conditions (imaginary movements of the left and right hand) in order to test the new training algorithm. The proposed algorithm outperformed the accuracy acquired by a dendrite ellipsoidal neuron based on k-means++ obtaining 76.02% and 62.77%, respectively. Also, the algorithm was compared with multilayer perceptrons and support vector machines which are some of the most common classifiers used to detect motor-related information in brain signals. These achieved an accuracy of 72.38% and 65.81%, respectively.
AB - Dendrite ellipsoidal neurons are a novel and different alternative for classification tasks, giving competitive results compared with typical classification methods. Based on k-means++ algorithm, the network allows each dendrite to build a hyperellipsoidal in order to assign each incoming pattern to its respective C class. The main disadvantage of this training algorithm is the lack of accuracy in high dimensional datasets. In this research, we solved this problem by training the dendrite ellipsoidal neuron using stochastic gradient descent. Furthermore, electroencephalography data were acquired during two mental conditions (imaginary movements of the left and right hand) in order to test the new training algorithm. The proposed algorithm outperformed the accuracy acquired by a dendrite ellipsoidal neuron based on k-means++ obtaining 76.02% and 62.77%, respectively. Also, the algorithm was compared with multilayer perceptrons and support vector machines which are some of the most common classifiers used to detect motor-related information in brain signals. These achieved an accuracy of 72.38% and 65.81%, respectively.
KW - Dendrite Ellipsoidal Neuron
KW - Electroencephalography
KW - Motor Imagery
KW - Multilayer perceptrons
KW - Stochastic Gradient Descent
KW - Support vector machines
KW - k-means++
UR - http://www.scopus.com/inward/record.url?scp=85068319051&partnerID=8YFLogxK
U2 - 10.1007/978-3-030-21077-9_8
DO - 10.1007/978-3-030-21077-9_8
M3 - Contribución a la conferencia
AN - SCOPUS:85068319051
SN - 9783030210762
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 80
EP - 88
BT - Pattern Recognition - 11th Mexican Conference, MCPR 2019, Proceedings
A2 - Carrasco-Ochoa, Jesús Ariel
A2 - Martínez-Trinidad, José Francisco
A2 - Olvera-López, José Arturo
A2 - Salas, Joaquín
PB - Springer Verlag
T2 - 11th Mexican Conference on Pattern Recognition, MCPR 2019
Y2 - 26 June 2019 through 29 June 2019
ER -