Skip to main content

Table 6 The time of data preprocessing + models training and testing of each model (second). Run on data science server at Chiang Mai University, Thailand (LINUX VPS, RAM 16 GB, CPU INTEL CORE i9, GPU 2080TI 11GB)

From: Influenza, dengue and common cold detection using LSTM with fully connected neural network and keywords selection

 

Dataset

dengue + cold

Flu + dengue + cold

Flu + cold

SMS Spam Collection Dataset

Model Architecture / Filing missing

LSTM

LSTM with numerical

LSTM + FNN

LSTM

LSTM with numerical

LSTM + FNN

LSTM

LSTM with numerical

LSTM + FNN

LSTM

original

 

13.3

17.7

8.6

11.9

15.7

10.1

26.3

27.6

12.7

98.5

MG

 

11.6

17.1

8.6

10.7

14.0

10.1

26.2

27.8

18.4

85.5

keywords selection (cut words: frequency < 2)

cubic interpolation

12.4

12.3

12.9

55.2

47.5

10.9

24.0

24.7

20.6

114.2

cut

13.4

17.6

8.6

10.9

17.1

11.7

26.3

27.6

12.5

107.7

mean

14.3

16.7

8.5

11.6

17.8

12.0

23.6

16.5

8.0

101.5

keywords selection (cut words: MI bottom 5%)

cubic interpolation

22.5

14.5

12.3

48.9

29.3

10.9

24.1

25.2

21.5

171.0

cut

22.9

15.6

11.5

12.5

19.3

10.2

24.9

25.7

12.8

111.0

mean

25.2

12.2

10.9

16.8

22.8

10.8

19.9

20.0

8.5

116.6

keywords selection (cut words: MI bottom 5% or frequency < 2)

cubic interpolation

13.4

14.7

13.0

49.0

29.3

11.0

24.8

25.3

20.0

111.6

cut

26.4

20.3

11.6

12.7

22.4

10.3

25.1

26.1

12.7

145.2

mean

25.1

12.6

10.7

22.1

22.2

10.8

21.7

19.9

8.1

105.3

keywords selection (cut words: frequency < 2) + MG

cubic interpolation

11.2

12.2

12.2

55.3

47.2

11.0

22.3

16.5

22.1

134.7

cut

11.5

16.9

8.7

10.6

13.8

10.2

26.6

27.6

18.1

87.5

mean

14.7

14.3

8.5

14.2

19.2

12.5

24.1

14.3

8.0

116.7

keywords selection (cut words: MI bottom 5%) + MG

cubic interpolation

23.9

15.1

11.1

27.6

28.9

10.9

24.0

25.3

20.0

206.7

cut

18.3

15.7

8.5

12.2

16.4

10.2

24.8

29.0

15.6

219.0

mean

25.0

11.8

10.2

16.6

20.6

11.1

15.4

16.7

15.4

116.8

keywords selection (cut words: MI bottom 5% or frequency < 2) + MG

cubic interpolation

13.4

13.6

11.0

36.0

29.5

10.8

23.9

25.4

18.4

147.3

cut

25.2

15.4

8.6

12.7

16.4

10.2

25.0

29.1

15.4

141.9

mean

19.7

11.8

8.5

15.8

20.6

10.9

21.6

17.0

8.0

80.2