Table 3 Comparison of the performance of the proposed model under different modal combinations.

From: A deep learning approach to stress recognition through multimodal physiological signal image transformation

 

Accuracy

Precision

Recall

F1-score

EDA + ACC

0.8220

0.8381

0.8359

0.8370

EDA + ACC + BVP

0.8616

0.8331

0.8307

0.8319

EDA + ACC + Resp

0.8249

0.8402

0.8359

0.8380

EDA + ACC + Temp

0.8418

0.8151

0.8151

0.8151

EDA + ACC + BVP + Resp

0.8729

0.8878

0.8828

0.8853

EDA + ACC + BVP + Temp

0.9096

0.9167

0.9167

0.9167

EDA + ACC + Resp + Temp

0.8757

0.8464

0.8464

0.8464

EDA + ACC + BVP + Resp + Temp

0.9068

0.9141

0.9141

0.9141

  1. Bold values indicate the best performance results among the compared methods for eachmetric.
Back to article page