Table 6 Performance analysis of the designed movement recognition model over state-of-the-art models.

From: Development of weighted residual RNN model with hybrid heuristic algorithm for movement recognition framework in ambient assisted living

Terms

SNN33

Bi-conventional RNN33

DBN43

LSTM36

HRS-COA-WRRNN

Accuracy

87.29

88.03

89.53333333

91.06111111

93.50111111

Sensitivity

78.45120942

79.6667181

82.08806287

84.58678597

88.58850222

Specificity

90.3145292

91.73548051

91.65462957

92.58814644

93.11971315

Precision

85.3

86.05333333

87.74666667

89.49

92.40666667

FPR

7.6854708

7.26451949

6.345370434

5.411853556

3.880286848

FNR

21.54879058

20.3332819

17.91193713

15.41321403

11.41149778

NPV

88.285

89.01833333

90.42666667

91.84666667

94.04833333

FDR

14.7

13.94666667

12.25333333

10.51

7.593333333

F1-Score

81.73238155

82.73696018

84.82309725

86.96933867

90.4573116

MCC

0.721616025

0.737248514

0.769484161

0.802485208

0.855771509

FOR

11.67142857

10.925

9.632142857

8.128571429

5.971428571

PT

15.55199895

15.29765353

13.91926042

12.38106924

10.39513948

CSI

69.29580727

70.28547981

73.56411331

77.35930204

82.64420623

BA

85.47439588

86.10754434

87.82072124

89.74899386

92.36928447