Skip to main content

Table 6 Deep neural network

From: Recognition of cooking activities through air quality sensor data for supporting food journaling

 

Original predictions

After post-processing

TN

3988

698

FP

957

454

TP

549

505

FN

120

164

Accuracy (%)

80.82

66.06

Specificity (%)

80.65

60.59

Recall (%)

82.06

75.49

Precision (%)

36.45

52.66

\(F_1\) score (%)

50.48

62.04

  1. Results with cooking instance modality, delayed recognition
  2. Numbers in italic indicate the largest value obtained in the experiment for each considered metric
Back to article page