Have a personal or library account? Click to login
Classification of emotions based on electrodermal activity and transfer learning - a pilot study Cover

Classification of emotions based on electrodermal activity and transfer learning - a pilot study

Open Access
|Dec 2021

Figures & Tables

Figure 1

Example measurement of low frequency skin conductance. Person being highly stressed (red curve), moderately stressed (green curve), and totally relaxed (blue curve).
Example measurement of low frequency skin conductance. Person being highly stressed (red curve), moderately stressed (green curve), and totally relaxed (blue curve).

Figure 2

Conductance measurement. Example of data reported as disgust.
Conductance measurement. Example of data reported as disgust.

Figure 3

CWT treated conductance measurement. The same example (Figure 2) of data reported as disgust with CWT applied.
CWT treated conductance measurement. The same example (Figure 2) of data reported as disgust with CWT applied.

Figure 4

Architecture of the machine learning process.
Architecture of the machine learning process.

Figure 5

Confusion matrix for the first results produced by the model using the full dataset.
Confusion matrix for the first results produced by the model using the full dataset.

Figure 6

Confusion matrix for the first results produced by the model using the full data set with SMOTE data added.
Confusion matrix for the first results produced by the model using the full data set with SMOTE data added.

Figure 7

Confusion matrix for the results produced by the model using the three category data set with SMOTE data added.
Confusion matrix for the results produced by the model using the three category data set with SMOTE data added.

Results using the full CWT applied EDA dataset with synthetic data_ The parameters are set to: test size = 20 and trees = 100_

test size = 20trees = 100

precisionrecallf1-scoresupport
amusement 1.000.500.674
anger 0.600.750.674
disgust 0.330.250.294
fear 1.001.001.006
neutral 0.800.440.579
sadness 0.270.600.375

accuracy 0.5932
macro avg 0.670.590.5932
weighted avg 0.700.590.6132

Results using the 3 class CWT EDA data with synthetic data_ Test size = 20 and trees = 300_

test size = 20trees = 600

precisionrecallf1-scoresupport
amusement 1.000.800.895
disgust 0.800.670.736
sadness 0.711.000.835

accuracy 0.8116
macro avg 0.840.820.8216
weighted avg 0.840.810.8116

Results using the CWT treated EDA data with test size = 20 and trees = 100_

test size = 20trees = 100

precisionrecallf1-scoresupport
amusement 0.750.750.754
anger 0.000.000.002
disgust 1.000.250.408
fear 0.000.000.001
neutral 0.000.000.002
sadness 0.150.670.253

accuracy 0.3520
macro avg 0.320.280.2320
weighted avg 0.570.350.3520

Category of emotion and how many times the emotion was reported_

EmotionNumber of samples
Amusement20
Anger9
Disgust24
Fear5
Neutral15
Sadness26
Tenderness1
Language: English
Page range: 178 - 183
Submitted on: Dec 7, 2021
Published on: Dec 30, 2021
Published by: University of Oslo
In partnership with: Paradigm Publishing Services
Publication frequency: 1 issue per year

© 2021 Fredrik A. Jacobsen, Ellen W. Hafli, Christian Tronstad, Ørjan G. Martinsen, published by University of Oslo
This work is licensed under the Creative Commons Attribution 4.0 License.