Skip to main content

Table 14 Comparison with state-of-the-art methods on dataset1 with respect to accuracy

From: CAPHAR: context-aware personalized human activity recognition using associative learning in smart environments

Subjects

DeepSense

DRBLSTM

DeepConvLSTM

Wen et al. (Loc)

Wen et al. (Time)

AL

Subject 1

58.73%

59.45%

62.04%

59.27%

56.36%

64.91%

Subject 2

59.26%

63.69%

67.24%

61.43%

60.37%

69.39%

Subject 3

58.94%

62.37%

66.15%

62.23%

60.97%

67.71%

Subject 4

63.87%

67.65%

69.43%

65.52%

61.26%

73.91%

Subject 5

45.44%

48.59%

54.36%

46.20%

44.76%

54.52%

Subject 6

64.91%

68.84%

72.45%

68.26%

63.72%

72.00%

Subject 7

65.54%

69.42%

72.67%

65.36%

63.68%

72.54%

Avg

59.53%

62.86%

66.33%

61.18%

58.73%

67.85%

Avg E.T.T

2.04 s

1.45 s

1.73 s

0.24 s

0.27 s

0.32 s

  1. The best accuracy is represented by boldface
  2. *E.T.T, Execution Time for Testing