Table 5 summary of comparative studies

From: Evaluation of advanced Kalman filter on real-time agricultural soil parameters through an IoT resources-constrained device

Models

Parameters

Metrics

References

MSE

RMSE

R2

Random Forest

Nitrogen

0.99

41

Phosphorus

0.99

Humidity

0.99

Potassium

0.99

Temperature

0.99

pH

0.99

Support vector machine regression

Organic carbon

0.66

0.13

29

pH

0.58

0.36

Nitrogen

0.40

12.36

Phosphorus

0.71

49.52

Electric Conductivity

0.59

0.21

Potassium

0.24

26.87

Zinc

0.84

0.39

Sand

0.66

7.75

Silt

0.54

4.07

Clay

0.68

4.85

Partial Least Squares

Regression

Organic carbon

0.63

0.14

pH

0.61

0.34

Nitrogen

0.36

12.55

Phosphorus

0.72

48.09

Electric Conductivity

0.63

0.19

Potassium

0.13

28.55

Zinc

0.79

0.44

Sand

0.60

8.36

Silt

0.41

4.60

Clay

0.66

5.02

Autoregressive with Kalman filters

Soil temperature

0.40

27

Humidity

0.20

  1. R2—coefficient of determination, MSE—mean square error, RMSE—root mean Square Error, CM—Computation Memory (Mb), and CT—Computation Time (s).